Upload
others
View
5
Download
0
Embed Size (px)
Citation preview
P l a t o n i c C r a f t a n d M e d i c a l E t h i c s
by
Daniel William Bader
A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy
Philosophy Department University of Toronto
© Copyright by Daniel William Bader 2010
ii
Platonic Craft and Medical Ethics
Daniel William Bader
Doctor of Philosophy
Philosophy Department
University of Toronto
2010
Abstract
Platonic Craft and Medical Ethics examines the Platonic theory of craft and
shows its application to different ethical problems in medicine, both ancient and modern.
I begin by elucidating the Platonic use of the term ―craft‖ or ―technē‖, using especially
the paradigmatic craft of medicine, and explicate a number of important principles
inherent in his use of the term. I then show how Plato‘s framework of crafts can be
applied to two ancient debates. First, I show how Plato‘s understanding of crafts is used
in discussing the definition of medicine, and how he deals with the issue of ―bivalence‖,
that medicine seems to be capable of generating disease as well as curing it. I follow this
discussion into Aristotle, who, though he has a different interpretation of bivalence, has a
solution in many ways similar to Plato‘s. Second, I discuss the relevance of knowledge
to persuasion and freedom. Rhetors like Gorgias challenge the traditional connections of
persuasion to freedom and force to slavery by characterizing persuasion as a type of
force. Plato addresses this be dividing persuasion between sorcerous and didactic
persuasion, and sets knowledge as the new criterion for freedom. Finally, I discuss three
modern issues in medical ethics using a Platonic understanding of crafts: paternalism,
conclusions in meta-analyses and therapeutic misconceptions in research ethics. In
iii
discussing paternalism, I argue that tools with multiple excellences, like the body, should
not be evaluated independently of the uses to which the patient intends to put them. In
discussing meta-analyses, I show how the division of crafts into goal-oriented and causal
parts in the Phaedrus exposes the confusion inherent in saying that practical conclusions
can follow directly from statistical results. Finally, I argue that authors like Franklin G.
Miller and Howard Brody fail to recognise the hierarchical relationship between medical
research and medicine when they argue that medical research ethics should be
autonomous from medical ethics per se.
iv
Acknowledgements
I would like to gratefully acknowledge the assistance of the following:
My thesis supervisor, Rachel Barney, who has assisted me with this thesis from
the first stage to the last.
My thesis committee members, Brad Inwood, Joseph Boyle and Gopal
Sreenivasan, whose insightful comments helped me to shape my arguments.
Margaret Morrison, whose seminar helped me to develop the paper that turned
into Chapter Four, Section 2.
The Department of Philosophy at the University of Toronto, which provided a
welcoming and helpful environment in which to study.
The Collaborative Programme in Ancient and Medieval Philosophy at the
University of Toronto, which provided me with essential training.
The staff of the John M. Kelly Library, who helped me to find obscure resources
and provided me with workspace.
The School of Graduate Studies at the University of Toronto, whose funding
allowed me to continue my work.
My parents, Edward Bader and Carole Sinclair, who have provided me with years
of support.
My wife, Katherine Bader, who proofread every incarnation of every chapter of
this dissertation. Her help and encouragement have been indispensible.
v
Table of Contents
Abstract .............................................................................................................................. ii
Acknowledgements .......................................................................................................... iv
Introduction ....................................................................................................................... 1
1 Purpose ......................................................................................................................... 1
2 ―Technē‖ or ―Craft‖ ..................................................................................................... 3
3 Philosophical Issues ..................................................................................................... 4
4 Interpretive issues ........................................................................................................ 5
5 Chapters ....................................................................................................................... 7
Chapter 1: Plato on Crafts ............................................................................................... 9
1 Introduction .................................................................................................................. 9
2 Texts ............................................................................................................................. 9
3 Plato‘s Use of ―Technē‖ Compared to Aristotle‘s Definition .................................... 10
4 Productive and Acquisitive Crafts ............................................................................. 14
5 Crafts as Knowledge .................................................................................................. 21
5.1 The Co-Extensivity of ―Technē” and ―Epistēmē‖ ........................................................... 21
5.2 The Teachability of the Crafts ......................................................................................... 23
6 Crafts and Dialectic.................................................................................................... 25
7 Crafts and Measurement ............................................................................................ 35
8 The Hierarchy of Crafts ............................................................................................. 43
9 Chapter Conclusion .................................................................................................... 51
Chapter 2: Crafts and Contraries ................................................................................. 53
1 Introduction ................................................................................................................ 53
2 Texts ........................................................................................................................... 54
3 Plato on the Problem of Crafts and Contraries .......................................................... 55
4 Plato‘s Solution .......................................................................................................... 64
4.1 The Linguistic Space ....................................................................................................... 64
4.2 The Substantive Argument .............................................................................................. 70
4.3 Conclusion ....................................................................................................................... 76
5 Aristotle on the Problem of Crafts and Contraries ..................................................... 77
6 Aristotle‘s Solution .................................................................................................... 87
6.1 The Linguistic Space ....................................................................................................... 88
vi
6.2 The Substantive Argument .............................................................................................. 92
6.3 Aristotle on Disease as a Privation .................................................................................. 99
6.4 Conclusion ..................................................................................................................... 102
7 Chapter Conclusion .................................................................................................. 103
Chapter 3: Plato on Persuasion and Knowledge ........................................................ 104
1 Introduction .............................................................................................................. 104
2 The Texts ................................................................................................................. 105
3 Eleutheria, Freedom and Autonomy........................................................................ 107
4 The Traditional Synthesis ........................................................................................ 109
5 Gorgias‘ Challenge .................................................................................................. 117
5.1 The Historical Gorgias .................................................................................................. 117
5.2 Plato‘s Gorgias .............................................................................................................. 119
6 Plato‘s Response ...................................................................................................... 122
6.1 Persuasion and Magic .................................................................................................... 122
6.2 Knowledge and Freedom ............................................................................................... 127
7 Medical Applications ............................................................................................... 136
7.1 Sorcerous Persuasion in Medicine ................................................................................. 136
7.2 Didactic Persuasion in Medicine ................................................................................... 141
8 Chapter Conclusion .................................................................................................. 148
Chapter 4: Modern Applications ................................................................................. 150
1 Introduction .............................................................................................................. 150
2 Paternalism and Autonomy ...................................................................................... 151
2.1 The Current Paradigm ................................................................................................... 152
2.2 The Powers of the Body ................................................................................................ 154
2.3 Healthier and Unhealthier .............................................................................................. 159
2.4 Medical Decisions and Collaboration............................................................................ 163
2.5 Rethinking the Paradigm ............................................................................................... 167
3 Recommendations in Meta-Analyses ...................................................................... 169
3.1 Introduction ................................................................................................................... 169
3.2 The ―Hippocratic Method‖ and the Scientific Method .................................................. 170
3.3 Actual Cases .................................................................................................................. 172
3.4 Concern about Precision ................................................................................................ 176
3.5 Conclusion ..................................................................................................................... 179
vii
4 The Therapeutic Misconception .............................................................................. 179
4.1 Introduction ................................................................................................................... 179
4.2 The Therapeutic Misconception .................................................................................... 180
4.3 Conclusion ..................................................................................................................... 187
5 Chapter Conclusion .................................................................................................. 187
Conclusion ..................................................................................................................... 189
1 Summary .................................................................................................................. 189
2 Important Results ..................................................................................................... 190
3 Future Applications .................................................................................................. 191
Bibliography .................................................................................................................. 194
1
I n t r o d u c t i o n
1 Purpose
Medicine is referred to 377 times in the Platonic corpus, more even than politics
(219 references) or rhetoric (290 references), and is mentioned in every single Platonic
dialogue except for the Hippias Major and the Menexenus.1 Despite Plato‘s nearly
ubiquitous use of medicine, few have directly addressed what Plato had to say about it.
The origins of this dissertation lay in examining those 377 references to discover what it
was that Plato had to say about medicine. What I found was that behind Plato‘s use of
medicine is a rich theory not only of medicine, but of all the crafts. Plato‘s understanding
of medicine and his understanding of crafts were so closely intertwined that to understand
one was to understand the other. Medicine, like all crafts, is a hybrid of knowledge,
power and practical reason, and it is often by using the paradigmatic craft of medicine
that Plato examines the relationships between the three. In this dissertation, I focus on
those relationships, as well as the implications of these relationships for modern
discussions in medical ethics.
Much has been written about Plato and the crafts. Gregory Vlastos, Terry Penner,
Terrence Irwin and David Roochnik all provide extensive discussion of what is called the
―craft analogy‖. According to the craft analogy, virtue either is a craft or is so analogous
to a craft that it functions in largely the same way. That this would be the focus of
scholarship is not surprising. Evaluating the strengths and weaknesses of the craft
analogy is one of the main uses to which Plato puts his own discussion of crafts in his
dialogues, and it is a question addressed in almost exactly those terms by Aristotle in
Nicomachean Ethics VI.
However, this dissertation takes quite a different approach. Crafts are important
not only because of their analogy to virtue, but in their own right. Understanding the
crafts, their nature, their purpose and their relationships to each other and to other types
of knowledge is important to properly understanding professional ethics such as those of
medicine, law and even teaching. I focus especially on medicine for two reasons. First,
1 In order to generate this number, I searched for the roots, ―ηαηξ‖, ―πνιηηηθ‖, ―ξεηνξ‖ on University of
California, "Thesaurus Linguae Graecae" <http://www.tlg.uci.edu/> [accessed 22 September 2009].
2
it is perhaps the profession whose ethics are the most controversial and many of those
controversial discussions centre on debates about the proper relationship between
medicine, science and politics. Second, it is the craft that Plato uses as his paradigmatic
case of a craft.
What I discovered from my examination of Plato‘s understanding of crafts and
medicine is that crafts play a central role in Plato‘s understanding of human action. The
crafts are the point of intersection of knowledge and power. With a craft, knowing that
turns into knowing how, by the understanding of causal relationships and possible
results2. The ways in which knowledge of facts transforms itself into power is a central
concern of Plato‘s, and serves as part of his claim that knowledge is the source of
freedom in texts such as the Laws. Plato undifferentially treats crafts both as a type of
knowledge and as a type of power and many of the results of his arguments come from
his treating crafts as both simultaneously. Second, crafts are the point of intersection of
science and practical reason. They are where knowing how meets knowing when.
Understanding causal relationships provides the power to produce certain results, but one
still needs the understanding of when one should produce them. Given that crafts serve
as a nexus for knowing that, knowing how and knowing when, it is no surprise that
discussion of crafts serves a central role in Plato‘s discussions of the relationships
between knowledge, power and virtue.
An important component of Plato‘s understanding of crafts is not only their
hybrid nature, but also the relationship between the various crafts. Plato sees that crafts
were not autonomous bodies of knowledge that could be understood independently from
one another. Instead, crafts are connected in a hierarchical relationship. There would be
no flute-makers if there were no flautists and there would be no scalpel-making if there
was no medicine. Understanding the nature of a craft requires not only being able to
recognize the product and how the craft produces it, but the uses to which the product
will be put, something that comes from another craft. This hierarchical relationship
guides Plato‘s understanding of authority in the crafts and is directly applicable to
2 This is not intended to mirror Gilbert Ryle‘s strict distinction between ―knowing how and knowing that‖
presented in Gilbert Ryle, "Knowing How and Knowing That", Proceedings of the Aristotelian Society, 46
(1946): 1-16. Rather, the ―knowing how‖ of Plato‘s crafts is not distinguishable from the ―knowing that‖
knowledge of causes and the product.
3
discussions of the relationship between physicians and patients and between physicians
and medical research.
This dissertation, then, is about the Platonic understanding of crafts, mediated
through his use of the paradigm of medicine. I do not explore in detail the purposes to
which Plato puts his theory, that is, to virtue or politics. Rather, I examine his
understanding of the crafts themselves. As professional ethics, that is, the ethics of
craftspeople, becomes an expanding field of philosophy, it becomes increasingly
important to understand the nature of crafts themselves. I therefore examine Plato‘s
theory of the crafts directly.
2 “Technē” or “Craft”
As ―craft‖ is the primary term that this dissertation will discuss, it is important to
clarify what is meant by the term. I will cover the details of Plato‘s use of the term in
Chapter One, but here I will try to situate the word in English vocabulary. The term
translates the Greek word, ―technē‖ (pl. technai), a word that, like many Greek words,
does not map on perfectly to any English word. Its intension is basically, ―an expert body
of knowledge‖, though the ―expertise‖ here does not rule out the possibility that everyone
might share in the knowledge. The Greek-English Lexicon provides the following
meanings: ―an art, craft,...trade,...profession‖3. The difficulty is that the Greek word is
very broad, and no one English word is broad enough to capture all of the instances while
remaining sufficiently grammatically flexible. I will provide some possible translations
and describe the strengths and weakness of them:
1. ―Craft‖: This is the translation I have chosen to use. It has the advantage of
having all the right grammatical forms, and even captures the secondary sense of
―craftiness‖ and ―cunning‖. Its disadvantage is that the word ―craft‖ doesn‘t
capture non-productive ―technai‖ such as geometry or astronomy, and the word
has a quasi-trivial sense, associated with things like summer camp.
3 H.G. Liddell and R. Scott, Greek-English Lexicon with a Revised Supplement (Oxford: Clarendon Press,
1996), 1785.
4
2. ―Art‖: This is English word from the Latin ―ars‖, the word into which ―technē‖
was translated. It lacks any verb forms, and is easily confused with fine arts.
―Art‖ shares the limitation of ―craft‖ that the word implies productivity.
3. ―Expertise‖: This word probably comes closer to the meaning of the term
―technē‖, in that it captures both productive and theoretical ―technai‖. However,
it lacks a verb and, worse, a plural. Its primary intensional weakness is that it
implies a relative rather than an absolute meaning. Not everyone can be an
expert, while some ―technai‖, like speaking grammatically, almost everyone has.
4. ―Technology‖: While obviously etymologically connected to ―technē‖,
technology is too specific in its meaning. It shares the weaknesses of ―craft‖ in so
far as it is limited to production, while it shares the weaknesses of expertise in that
it lacks most grammatical forms. However, when trying to situate debates about
―technē‖ in a modern context, many of those debates would be described using
the term ―technology‖, such as the relationship of technology to science or of
technology to ethics. It is therefore useful for situating the ancient debates, but
functions poorly as a translation.
None of these words are perfect. ―Expertise‖ is probably closest in terms of intension,
but it is grammatically inflexible. ―Craft‖ is second-best in intension and best for
grammatical versatility, though one must be aware that ―technē‖ includes non-productive
activities like geometry. Therefore, I have chosen to translate the term ―technē‖ as
―craft‖.
3 Philosophical Issues
Plato‘s theory of crafts opens up different types of philosophical issues that
should be kept distinct, despite their overlap in practice. First, Plato‘s theory of crafts can
help settle definitional questions. In Chapters Two and Three, I show how resolving
these definitional questions was vital to understanding ancient debates around the craft
analogy and around freedom. Chapter Two, for example, is dedicated to addressing the
question of whether or not poisoning by a doctor should be considered medicine. In
Chapter Four, I show how resolving these definitional questions is vital to understanding
5
some contemporary biomedical debates. The second and third parts of Chapter Four
include discussion of the distinction between the craft of medicine and the knowledge of
causation that underlies it.
Second, Plato‘s theory of crafts can help resolve ethical discussions, especially
those in the professions. Two conclusions arise from the Platonic theory. First, crafts are
not independent entities. Most importantly, crafts that produce the tools of other crafts
need to be guided by the crafts that employ their products in several ways. Crafts are
both interdependent and incomplete. Flute-makers are not only dependent on flautists to
play their products, but they are dependent on flautists for the form of their products and
to evaluate the quality of their work. I will discuss this hierarchical relationship in detail
in Chapter One. Second, if a kind of activity is one that admits of expertise, that activity
will best be performed by someone with that expertise. Conversely, unless someone has
the relevant expertise in a case where expertise is required, one will be powerless. This
raises the question of what kind of knowledge someone ought to have in order to be ―in
charge‖ in a particular craft-related decision. This question is very important in
understanding the kind of knowledge that a person would need to have in order to act
freely in the strong sense of the word, that is, to act autonomously.4
4 Interpretive issues
As in any discussion of Plato, several interpretive issues must be discussed. After
all, Plato wrote dialogues, not treatises, and one must provide some summary of one‘s
approach to the dialogues. Moreover, certain interpretive problems arise as I am often
not using Plato‘s arguments to discuss the issues with which he was ultimately
concerned.
Discussing what Plato has to say about medicine often pits one against the
intention of the dialogues, which leads to some interpretive difficulties. Despite the
ubiquitous medical references, Plato does not appear to be interested in medicine per se.
There is no dialogue called ―The doctor‖, and there is no dialogue whose main theme is
medicine. Medicine is a ―paradigm‖ in the sense of the term used in the Euthyphro: it is
4 I discuss the relationship between the meanings of the Greek term ―eleutheria‖ (freedom) and the English
term ―autonomy‖ in Chapter Three.
6
something looked at for the sake of evaluating other cases. It is not, however, examined
for its own sake. This requires me to approach the texts obliquely. Quite often, what
Plato has to say about medicine often serve as premises for analogous arguments about
something else. It is the premise, after all. Because I am swimming against the current
of the argument, I do one of two things in order to better understand these premises:
1. If I can extrapolate from some other argument in another dialogue why Plato says
what he says about medicine, I will do so.
2. If I can find no source at all, I will simply report what Plato has to say and note
that I have done so.
Since Plato‘s discussion of crafts and medicine often serves in the role of premises for
other arguments, I will be required to examine what he has to say about medicine by
examining only parts of the arguments presented.
This approach requires me to jump from dialogue to dialogue quite a bit, as I
search for explanations as to why Plato might assume a certain premise for an argument
in another dialogue. This assumes a unitarian approach between the dialogues. If I were
to treat the various dialogues as completely independent from one another, the statements
from one dialogue would be irrelevant to the other. Fortunately, the presuppositions of
the arguments in Plato‘s dialogues are often far more unified than their conclusions. In
many dialogues in which the main characters disagree, they continue to agree on their
basic use of terms. For example, despite their strong disagreement about what kind of
craft constitutes justice, Thrasymachus and Socrates seem to be in agreement about what
constitutes a craft, and it is this very agreement that undermines Thrasymachus. Even
when Plato provides different sides of arguments, his premises are often the same, and it
is these that I am examining. Between dialogues, too, this same unity of presuppositions
is often stronger than the variation between the conclusions5.
On another interpretive question, I‘ve taken something of a controversial stance:
Plato‘s use of jokes. Often, Socrates and other interlocutors make quite comical
statements, and this leads to some reasonable doubt as to how seriously the reader is
expected to take the arguments in which they are present. In trying to understand Plato‘s
5 For example, Socrates appears to shift his position between the Phaedrus and the Gorgias, concerning
whether or not rhetoric is a craft. The characteristics of craft, however, stay largely the same.
7
approach to crafts, I often take portions of these jokes at face value. My reasoning is that
Plato quite often includes his serious premises and presuppositions in his more humorous
passages. In fact, the presuppositions are often a part of the ―set up‖ of the joke. For
example, in the Republic, Socrates argues that soldiers need to be trained in mathematics,
so that they can count the troops they have and not get lost. The passage seems quite
comical. However, a similar claim appears in the Philebus, in which Socrates argues that
all crafts need measurement and are only craft-like in as much as they include
measurement. This passage seems serious. Here is an example of Plato using a
presupposition as a part of humour, without it implying that we should therefore not take
the presupposition itself seriously. This approach to humour is very important to how I
approach the division passages in the Sophist and Statesman, which I take to provide
serious discussion of how to collect and divide, while at the same time making some
divisions that are deliberately humorous.
5 Chapters
The chapters in this dissertation introduce the Platonic understanding of crafts,
and then attempt to apply that understanding to various problems that arose in the ancient
world and arise in the modern one. Chapters Two through Four are not an exhaustive
survey of the kinds of problems that one could use the Platonic understanding of crafts to
discuss. Rather, they are examples of the kinds of applications that Plato‘s understanding
of crafts and the craft of medicine did and can have.
Chapter One includes a summary of Plato‘s understanding of crafts. Much of the
material here is uncontroversial. However, I have dedicated extra space to discussing
Plato‘s use of hierarchy and the special role that the understanding of causal relationships
plays in a craft. The discussion of hierarchy is important for understanding how the
various relationships between craftspeople ought to be established.
Chapter Two is a discussion of an important theoretical debate about the nature of
crafts that concerned both Plato and Aristotle. Considered as knowledge, crafts seem to
make people capable of contraries, that is, a good doctor is also a good poisoner.
However, considered as a power, crafts can only do one thing. Plato and Aristotle take
quite different routes and arrive at the same conclusion that medicine, strictly speaking,
8
can only heal. This debate has modern resonance in disscussions of the justification and
purpose of ethical rules such as the World Medical Association‘s 1948 Geneva
Declaration‘s precept that, ―The health of my patient should be my first consideration.‖ 6
Chapter Three is a discussion of Plato‘s discussion of the relationship between
freedom, power and knowledge. In the Gorgias and other dialogues, Plato raises
concerns that the democratic paradigm that to be free is to be persuaded rather than
forced does not provide sufficient protection for people from the manipulative rhetoric of
people like Gorgias. Instead, Plato puts forward a new criterion of freedom: knowledge.
This chapter examines the arguments he presents for this new criterion and the sorts of
knowledge that is required to be considered free. The term for freedom, ―eleutheria‖,
shares much in common with our modern understanding of autonomy, and Plato‘s
discussion of the knowledge that is necessary for freedom is important for understanding
modern discussions of autonomy in medicine.
In Chapter Four, I apply the Platonic framework directly to modern problems in
biomedical ethics. First, I will discuss the weakness of the current paradigm for the
discussion of paternalism and autonomy. Currently, the physician is treated as the expert
in health, while the patient can only contribute non-health goals that can only
compromise their health. I will argue that, when one applies Plato‘s hierarchy of crafts, it
becomes clear that medicine is incomplete, and that patient guidance is actually a part of
the medical craft, properly understood. I then discuss the difficulties with
recommendations in meta-analyses such as those in the Cochrane Review, and the kinds
of confusions that arise when one conflates understanding of causal relationships with
understanding of how to apply them. Finally, I look at the ―therapeutic misconception‖
in research ethics described by Franklin Miller and Howard Brody, arguing that, while
there are indeed numerous problems with confusing research and therapy, the two are so
closely intertwined that the former should not have autonomous ethical standards, as
Miller and Brody suggest.
6 World Medical Association, "Declaration of Geneva", in Australian Medical Association Website
<http://www.ama.com.au/node/2474> [accessed 14 April 2010].
9
C h a p t e r 1 : P l a t o o n C r a f t s
1 Introduction
Instead of dealing with medicine directly, Plato uses it almost exclusively as a
paradigmatic example of what he calls a ―craft‖ or ―technē‖ (pl. ―technai”).7 In order to
understand what Plato has to say about medicine, and then to discuss its relevance to both
ancient and modern medical ethics in future chapters, one must therefore explain what
Plato means by a ―technē‖. Fortunately, given both the wide use of medical examples
and other discussions of crafts per se, there is no shortage of evidence from which to
draw such conclusions. In this section, I will discuss several of the key elements of crafts
that Plato describes throughout his works. First, I will discuss the nature of crafts
themselves, showing how Plato uses the term ―technē‖ to describe any definite set of
knowledge surrounding a given subject matter. Second, I will discuss the different types
into which Plato divides crafts, especially his distinction between acquisitive crafts and
productive crafts, of which medicine is the latter. Third, I will discuss the crafts‘
relationship with two meta-crafts that are pertinent to any craft, dialectic and
measurement. Finally, I will discuss the relationships among the crafts, as Plato arranges
the crafts in a hierarchy based on which crafts use the products of which other crafts as
tools. In addressing these topics, I will highlight seven principles that I will refer to in
future chapters to discuss the application of Platonic use of ―technē‖ to both ancient and
modern medical ethical issues. These characteristics of and relationships between crafts
will provide the theoretical foundation that will guide the ethical conclusions I will
develop in the second through fourth chapters.
2 Texts
While Plato discusses the nature and relationships of technai in many of his
dialogues, in some dialogues he deals with the crafts in great detail, often in the context
7 I discuss the reasoning behind the choice of ―craft‖ as a translation of ―technē‖ in the introduction to the
dissertation.
10
of discussions of politics or rhetoric. As I primarily will be using these texts, I will
introduce them here:
Phaedrus (257c-279c): In an extended debate concerning whether or not rhetoric can
properly be considered a craft, Socrates provides and defends a long description of the
methodology of medicine and its relationship to dialectic.
Sophist (216a-237a): In trying to find the definition of a sophist, a putative craftsperson,
Socrates provides a number of ways of classifying the crafts, defining medicine as a
productive rather than as an acquisitive craft.
Statesman (passim): In a detailed search for the political craft, Socrates considers several
important questions about the nature of crafts, including whether therapeutic crafts such
as medicine should be divided according to their subjects or their methods, and how
different sorts of measurement are relevant to the various crafts.
Philebus (23c-27c): This dialogue provides an extended discussion of generation by
productive crafts. Socrates provides a further discussion of the art of measurement and
its relationship to the products of crafts.
3 Plato’s Use of “Technē” Compared to Aristotle’s
Definition
Unfortunately, Plato never provides a clear definition of a craft in any of his
dialogues. Despite 774 uses of ―technē‖ or one of its derivatives in the Platonic corpus,
at no point does any interlocutor provide a clear definition of what a craft is.8 This does
not, however, stop Plato from using the term regularly nor does it stop him from making
claims about crafts. Instead of speaking of Plato‘s ―definition‖ of craft, then, I will
discuss his ―use‖ of the term.9 For discussing his use of the term, Aristotle‘s clear
8 In order to generate this number, I searched for the root ―ηερλ‖ on University of California, "Thesaurus
Linguae Graecae" <http://www.tlg.uci.edu/> [accessed 22 September 2009]. 9 I do not wish to take quite the strong position of John Lyons, that we ought ―...to accept everything that
the native speaker says in his language, but to treat with reserve anything he says about his language, until
11
definition in the Ethics10
and Metaphysics provides an extremely helpful starting point for
four reasons: first, as Plato‘s student, he uses some of the same terms as Plato, helping to
illuminate Plato‘s usage; second, Aristotle does, in fact, provide a clear definition of
―technē‖ that can serve as a starting point for discussion of Plato‘s use; third, Aristotle‘s
definition of ―technē‖ has been sufficiently influential that it is often the conception that
one brings to the discussion of Plato; finally, Plato‘s usage is sufficiently different from
Aristotle‘s definition that the differences between the former‘s usage and the latter‘s
definition highlights what is distinctive about Plato‘s usage. Starting with Aristotle‘s
definition of ―technē‖ is not an uncommon approach to Plato‘s dialogues; indeed,
Aristotle‘s definition is of such clarity that it has been used in the discussion of Platonic
dialogues in preference to Plato‘s own usage, such as by Terence Irwin in his book
Plato’s Ethics:
It is useful, then, to replace the rather imprecise question ‗Does
Socrates teach virtue as a craft?‘ with the more precise question
‗Does Socrates treat virtue as the sort of thing that Aristotle
regards as a craft?‘11
As my intention in this chapter is to elucidate Plato‘s usage of ―technē‖ itself rather than
discuss Plato‘s ethical theory of virtue, I will not be following this approach. Rather, I
will demonstrate how Plato‘s usage of the term ―craft‖ is significantly broader than
Aristotle‘s definition. In fact, Plato‘s use is broad enough that it is ultimately co-
extensive with what he calls knowledge or ―epistēmē‖.
Aristotle defines ―technē‖ in the Ethics, and describes its genus and differetia in
the Metaphysics. They both have different contexts. In the Ethics, Aristotle defines a
craft as a ―productive disposition with a true account‖ (1140a9-10).12,13
The context of
this has been checked‖ (John Lyons, Structural Semantics: An Analysis of Part of the Vocabulary of Plato
[Oxford: Basil Blackwell, 1963], 140). Rather, absent a precise definition, we are left with examining the
use of the term and the claims he makes about them. I will still give special attention to the claims about
crafts, as they represent Plato‘s reflection on the subject. 10
As the definition appears in a book common to the Nicomachean Ethics and the Eudemian Ethics, I will
simply refer to the text as the ―Ethics‖, avoiding taking a stand on the book‘s original context. 11
Terence Irwin, Plato's Ethics (Oxford: Oxford University Press, 1995), 70. 12
ἕμηο κεηὰ ιόγνπ ἀιεζνῦο πνηεηηθή. 13
For translations of Aristotle, I will be using Aristotle, The Complete Works of Aristotle: the Revised
Oxford Translation, ed. Jonathan Barnes (Princeton: Princeton University Press, 1984). However, the
translations in those volumes do not use standardized terms (for example, they may not use the same term
to translate ―technē” consistently) and occasionally insert terms that do not appear in the Greek. Therefore,
I often modify the original translations.
12
this definition is Aristotle‘s distinction between production and acting, limiting crafts to
production:
All craft is concerned with generation, i.e. with crafting and
theorizing how something may generate that is capable of either
being or not being, and whose principle is in the maker and not
in the things made. (1140a10-13)14
Aristotle uses this definition to argue that virtue should not be considered a craft, as it
doesn‘t produce anything, but is rather a sort of disposition. In the Metaphysics, he
categorizes craft as a rational, productive power. All powers can be divided into rational
and irrational, and what differentiates them is whether or not they include an account:
Clearly some potentialities will be non-rational and some will be
accompanied by an account. This is why all crafts, i.e all
productive forms of knowledge are potentialities. (1046b3-5)15
There are some differences between these passages. The first is a definition, while the
second is a categorization. In the Ethics, craft is defined as a ―disposition‖ (―hexis‖),
while in the Metaphysics, it is described as power (―dynamis‖). In both cases, reason is
relevant. In the Ethics, craft is said to be ―with a true account‖ (―meta logou alēthous‖),
while in the Metaphysics, the word ―true‖ is not used (―meta logou‖). In both cases,
crafts are said to be ―productive‖ (―poētikē‖).
A word should be said here about ―production‖, given the slightly different
semantic field covered by the English word ―production‖ and the Greek word ―poēsis‖.
The English noun ―production‖ is generally limited to artifacts, like beds, or substances,
like plants. Instead, for Aristotle, production includes the production of properties in
objects that already exist. For instance, in the Metaphysics passage, Aristotle says that
medicine can produce health, which is clearly not a substance or an artifact, but a
property in an already existing substance. In order to distinguish these sorts of
production, as they will be relevant in this and future chapters, I will refer to the creation
of artifacts and substances as ―essential production‖ and the creation of properties in
already existing objects as ―accidental production‖.
14
ἔζηη δὲ ηέρλε πᾶζα πεξὶ γέλεζηλ θαὶ ηὸ ηερλάδεηλ θαὶ ζεσξεῖλ ὅπσο ἂλ γέλεηαί ηη ηλ ἐλδερνκέλσλ θαὶ
εἶλαη θαὶ κὴ εἶλαη, θαὶ ὧλ ἀξρὴ ἐλ ηῶ πνηνῦληη ἀιιὰ κὴ ἐλ ηῶ πνηνπκέλῳ. 15
δῆινλ ὅηη θαὶ ηλ δπλάκεσλ αἱ κὲλ ἔζνληαη ἄινγνη αἱ δὲ κεηὰ ιόγνπ· δηὸ πᾶζαη αἱ ηέρλαη θαὶ αἱ
πνηεηηθαὶ ἐπηζηῆκαη δπλάκεηο εἰζίλ.
13
Aristotle‘s definition of ―technē‖ is useful for clarifying Plato‘s use of the term, in
part because Plato uses the term differently. Plato‘s use is broader than Aristotle‘s in two
important ways. First, Plato does not limit the crafts to those that are productive; rather,
he includes many activities that would fall outside of Aristotle‘s definition as crafts.
Second, in both the Ethics and the Metaphysics, Aristotle treats craft as a species of
power whose differentia is a (true) account. Plato, on the other hand, often treats crafts as
a species of knowledge.16
With a few important exceptions that I will note later, if
something was commonly believed to be a craft at the time, the use of ―technē‖ to
describe it is left unchallenged in the dialogues, and the conversation continues as though
the use of the term is unproblematic. Plato‘s use then is largely conventional, accepting
as a craft whatever was normally considered a craft. To give a sense of the breadth of
Plato‘s use of the term ―craft‖ throughout the corpus, the following is a list of all of the
activities explicitly referred to as a ―technai‖ by one or more of Plato‘s interlocutors:17
speechwriting lyremaking cooking? midwifery measurement
imitation hiring music lawmaking herding
moneychanging wool-carding befriending appropriation house-building
householding naming tool-making love wrestling
teaching persuading molding pottery braiding war
politics burning reciting poems? rhetoric? sophistry?
fishing deception slinging subduing kingship
farming knowing grammar writing gymnastics
separating dialectic image-making copying mercantilism
spinning cosmetics? arithmetic charioteering hunting
nourishing medicine heraldry fulling flattery?
punishment embellishment discerning reasoning dancing
16
I will discuss Plato‘s use of ―technē‖ both as knowledge and as power in Chapter Two. 17
The following is a much condensed version of Anne Balansard‘s sixty-page list of all craft-terms in Plato
from Anne Balansard, Technè dans les Dialogues de Platon: L’empreinte de la sophistique (Sankt
Augustin: Academia-Verl., 2001), 324-84.
14
The few cases in which the craft-status of a putative craft is challenged are marked with a
―?‖. Both of Aristotle‘s definitions shared the term ―productive‖, but notice from the
above list that not all of the crafts listed can be considered productive. Plato includes
music, politics, hunting, dialectic and measurement in his list of crafts, none of which
either creates artifacts or produces anything in some other object.
While Aristotle‘s definition of ―technē” has been extremely influential, it does not
capture the all of the uses of the term in Plato‘s dialogues. Plato uses the term in an
extremely free way, rarely challenging the status of a putative craft. For the rest of this
chapter, I will be discussing this broader, Platonic use of the term ―technē”, highlighting
as I go the various principles of crafts that recur throughout Plato‘s uses of the term.
4 Productive and Acquisitive Crafts
Principle #1: There are two sorts of crafts: productive and acquisitive.
As I have shown above, Plato uses the term ―crafts‖ in a broader way than
Aristotle. Plato, in fact, pre-emptively denies Aristotle‘s claim that all crafts are
productive. Instead, productive crafts are only a subset of the crafts, which include all
kinds of knowledge. While Plato uses the term ―crafts‖ in a broad, conventional way, he
explicitly subdivides the crafts into separate kinds. Plato subdivides the crafts in four
dialogues: the Charmides, Euthydemus, Sophist and Statesman. These divisions in these
dialogues are similar in many ways. He divides the crafts into so-called productive crafts
(―poētikē‖) and acquisitive (―thēreutikē” or ―ktētikē‖) crafts.18
The productive crafts are
in many ways similar to what Aristotle will call a ―craft‖. Acquisitive crafts, however,
are quite distinct. Acquisitive crafts include such apparently unrelated activities as war,
seduction, hunting and geometry.19
Among the acquisitive crafts, Plato includes what
18
There is a terminological difference between the Euthydemus and the Sophist. In the former, these are
called ―thēreutikē‖ (―hunting‖), while in the latter they are called ―ktētikē‖ (―capturing‖) (For instance,
―ζεξεπηηθνὶ γάξ εἰζη θαὶ νὗηνη‖ at Euthydemus 290c and ―ηέρλε ηηο θηεηηθὴ‖ at Sophist 219c). I have
chosen the term ―acquisitive‖ to mark their overlap. 19
David Roochnik chooses to call this latter group of crafts ―theoretical‖ rather than ―acquisitive‖.
However, this creates too much possible confusion between Aristotle‘s distinction between theoretical and
practical knowledge and Plato‘s ―theoretical‖ and productive crafts. While it is strange to call geometry
―acquisitive‖, it is even stranger to call hunting ―theoretical‖. In the Euthydemus, Clinias explains why we
15
Aristotle later will call theoretical knowledge, including dialectic and mathematics20
.
However, Plato defends this category, arguing that acquisitive crafts represent a distinct
category of crafts.
Plato confronts an Aristotelian definition of crafts in the Charmides, in which
Critias raises an objection to Socrates‘ surprisingly Aristotelian use of the term ―technē‖.
Socrates attempts to argue against Critias‘ definition of temperance by assuming, like
Aristotle, that all crafts are necessarily productive:
And if you should ask me about housebuilding, which is a
knowledge of building houses, and ask what product I say that it
produces, I would say that it produces houses, and so on with the
other arts. So you ought to give me an answer on behalf of
temperance. (165d)21,22
Socrates‘ argument here assumes that all crafts are productive, so temperance should also
be productive. Therefore, Critias ought to provide the product of temperance. However,
Critias does not respond to Socrates in the way that one might expect. Rather than either
providing the product of temperance or denying that temperance is a craft, he instead
explicitly denies that all crafts are productive:
For instance…in the crafts of calculation and geometry, tell me
what is the product corresponding to the house in the case of
house-building and the cloak in the case of weaving and so on –
one could give many instances from many crafts. (165e-166a)23
Critias counters that there are, in fact, many crafts that do not have products, and
specifies the mathematical crafts of calculation and geometry. If Plato were committed to
the Aristotelian definition, one would expect that Socrates would either try to show that
geometry was somehow productive or deny that geometry is a craft. Importantly, he does
neither. Rather, Socrates openly concedes the point, saying ―you are right‖, and allows
ought to take the claim that geometry is acquisitive literally (David Roochnik, "Socrates' Use of the
Techne-Analogy", Journal of the History of Philosophy, 24 [1986]: 296). 20
See, for example, Metaphysics E:1. 21
Καὶ εἰ ηνίλπλ κε ἔξνην ηὴλ νἰθνδνκηθήλ, ἐπηζηήκελ νὖζαλ ηνῦ νἰθνδνκεῖλ, ηί θεκη ἔξγνλ ἀπεξγάδεζζαη,
εἴπνηκ‘ ἂλ ὅηη νἰθήζεηο· ὡζαύησο δὲ θαὶ ηλ ἄιισλ ηερλλ. ρξὴ νὖλ θαὶ ζὲ ὑπὲξ ηῆο ζσθξνζύλεο. 22
Translations from Plato will use as a baseline those from Plato, Plato: Complete Works, ed. John Cooper
(Indianapolis: Hackett, 1997). However, I will change the vocabulary and occasionally the phrasing, so
that that it matches the standard vocabulary in the dissertation. 23
ἐπεὶ ιέγε κνη, ἔθε, ηῆο ινγηζηηθῆο ηέρλεο ἠ ηῆο γεσκεηξηθῆο ηί ἐζηηλ ηνηνῦηνλ ἔξγνλ νἷνλ νἰθία
νἰθνδνκηθῆο ἠ ἱκάηηνλ ὑθαληηθῆο ἠ ἄιια ηνηαῦη‘ ἔξγα, ἃ πνιιὰ ἄλ ηηο ἔρνη πνιιλ ηερλλ δεῖμαη;
16
for a broader application of the term craft, arguing instead, that while not all crafts have
products, they nonetheless have subjects:24
You are right. But I can point out to you in the case of each of
these bodies of knowledge25
what it is knowledge of, this being
distinct from the knowledge itself. For instance, the craft of
calculation, of course, is of the odd and even…isn‘t that so?
(166a)26
After this response, both Critias and Socrates continue the discussion, agreeing that not
all crafts do have products, contrary to the Aristotelian definition.
However, this raises an important problem. If the subjects of non-productive
crafts are not products, what is their relationship to their crafts? The answer is not given
in the Charmides. However, in the Euthydemus, the young Clinias provides a description
of the relationship between mathematical crafts and their subjects, and the answer is in
some ways quite surprising. The relationship is analogous to that between hunter and
prey:
No craft of actual hunting, he said, extends any further than
pursuing and capturing: whenever the hunters catch what they
are pursuing they are incapable of using it, but they and the
fishermen hand over their prey to the cooks. And again,
geometers and astronomers and calculators (who are hunters,
too, for none of these make their diagrams; they simply discover
those which already exist), since they themselves have no idea of
how to use their prey but only how to hunt it, hand over the task
of using their discoveries to the dialecticians… (290b-c)27
24
In Plato’s Ethics, Terence Irwin argues that Socrates does not intend to imply in the Charmides that
geometry lacks a product. Rather, he argues that the correct answer that arises from calculation is the
product of those calculations. However, if this interpretation is correct, it is difficult to see what the point
of disagreement between Critias and Socrates is. Critias denies explicitly that a number of arts have
product, and Socrates says, ―You are right‖. When he starts talking about what crafts are ―of‖, the term
―products‖ disappears completely. Irwin, however, implies that the term ―product‖ appears in 166a, saying
―he insists that every science must have a product that is 'something other than the science itself‘ (166a3-
7)‖, though the word ―product‖ does not actually appear in the text quoted and Critias has just denied their
presence in a number of crafts, with Socrates‘ agreement (Terence Irwin, Plato's Ethics [Oxford: Oxford
University Press, 1995], 71). 25
The term “epistēmē”, which I will translate as ―knowledge‖, creates the difficulty that ―epistēmē‖ has a
plural (“epistēmai‖), while ―knowledge‖ does not. For the term ―epistēmai‖, I will use the phrase ―bodies
of knowledge‖. Please note that there is no corresponding word in Greek for ―bodies‖. The alternative
translation, ―science‖, has a plural but lacks a verb, which is worse. 26
Ἀιεζῆ ιέγεηο· ἀιιὰ ηόδε ζνη ἔρσ δεῖμαη, ηίλνο ἐζηὶλ ἐπηζηήκε ἑθάζηε ηνύησλ ηλ ἐπηζηεκλ,
ὃ ηπγράλεη ὂλ ἄιιν αὐηῆο ηῆο ἐπηζηήκεο. νἷνλ ινγηζηηθή ἐζηίλ πνπ ηνῦ ἀξηίνπ θαὶ ηνῦ πεξηηηνῦ, πιήζνπο
ὅπσο ἔρεη πξὸο αὑηὰ θαὶ πξὸο ἄιιεια· ἤ γάξ; 27
Οὐδεκία, ἔθε, ηῆο ζεξεπηηθῆο αὐηῆο ἐπὶ πιένλ ἐζηὶλ ἠ ὅζνλ ζεξεῦζαη θαὶ ρεηξώζαζζαη· ἐπεηδὰλ δὲ
ρεηξώζσληαη ηνῦην ὃ ἂλ ζεξεύσληαη, νὐ δύλαληαη ηνύηῳ ρξῆζζαη, ἀιι‘ νἱ κὲλ θπλεγέηαη θαὶ νἱ ἁιηῆο ηνῖο
17
This is a peculiar passage for a number of reasons. One must ask just how literally
Clinias expects his hearers to take the comparison of hunting and geometers28
. Clinias
appears to take the comparison literally, even providing us with reasons to think that they
are the same in important respects. First, geometers and hunters do not create anything;
like hunters, their crafts have no products. However, this is a negative property and does
not tell us what an acquisitive craft‘s relationship to its subject is. Second, then, Clinias
tells us what exactly the relationship is. Geometers, like hunters, discover their objects
through hunting.29
The definition of an acquisitive craft is a craft that finds and obtains
its subject; the relationship of an acquisitive craft to its subject is like that of hunter to
prey. There appears to be an important dissimilarity between hunting and geometry,
though, in terms of what they do with the subject once they have discovered it. The
geometer does not capture his or her subject in the same way as a hunter does; the word
―capture‖ in the first part of the analogy is replaced with the word ―discover‖ in the
second part of the analogy. Further, unlike a hunter, a mathematician does not alter the
prey in any way, as, aside from a few types of hunting using non-harmful traps, prey in
hunting tends to come out worse for wear.30
Another important aspect of this division is that Plato goes out of his way to
highlight this distinction to the reader. The dialogue Euthydemus is a report of a
conversation between Socrates, Euthydemus, Ctesippus, Dionysodorus, and Clinias by
Socrates to Crito. Right after the comparison of hunters and mathematicians quoted
ὀςνπνηνῖο παξαδηδόαζηλ, νἱ δ‘ αὖ γεσκέηξαη θαὶ νἱ ἀζηξνλόκνη θαὶ νἱ ινγηζηηθνί—ζεξεπηηθνὶ γάξ εἰζη θαὶ
νὗηνη· νὐ γὰξ πνηνῦζη ηὰ δηαγξάκκαηα ἕθαζηνη ηνύησλ, ἀιιὰ ηὰ ὄληα ἀλεπξίζθνπζηλ—ἅηε νὖλ ρξῆζζαη
αὐηνὶ αὐηνῖο νὐθ ἐπηζηάκελνη, ἀιιὰ ζεξεῦζαη κόλνλ, παξαδηδόαζη δήπνπ ηνῖο δηαιεθηηθνῖο θαηαρξῆζζαη
αὐηλ ηνῖο εὑξήκαζηλ, ὅζνη γε αὐηλ κὴ παληάπαζηλ ἀλόεηνί εἰζηλ. 28
Rosamond Kent Sprague in her translation, for example, takes it to be a simile,even adding the phrase ―in
a way‖ after ―who are hunters, too‖ (Plato, Plato: Complete Works, ed. John Cooper [Indianapolis: Hackett,
1997], 728). David Roochnik considers hunting to simply be a metaphor used to describe theoretical
knowledge: ―Socrates uses the metaphor of hunting to describe it‖ (David Roochnik, "Socrates' Use of the
Techne-Analogy", Journal of the History of Philosophy, 24 [1986]: 298). There is no need to reduce the
Euthydemus passage (and the Sophist passage) to mere metaphor, however, if theoretical knowledge is
simply a subset of acquisition. 29
There is something puzzling about Socrates‘ claim that geometers do not create their figures. In one
sense, of course they do. Each individual figure is drawn by the geometer. However, the claims they make
are not about the images, but that of which they are images: ―They make their claims for the sake of the
square itself and the diagonal itself, not the diagonal they draw, and similarly with the others‖ (Republic
510d). Perhaps the figures are to geometry what traps are to hunting and words are to learning. 30
Strictly speaking, this may not be a weakness in the analogy. The hunter per se only captures the prey,
and killing the prey is not a part of the definition of hunting.
18
above, the reporting of the conversation is suddenly broken, and Crito asks Socrates,
―What do you mean, Socrates? Did that boy utter all this?‖ (290d).31
Curiously, Socrates
becomes confused as to whether or not it was Clinias who actually made the comparison.
Instead, Socrates says that it may not actually have been Clinias, but rather some sort of
superior being (―tis tōn kreittonōn parōn”): ―Do you suppose, my good Crito, that some
superior being was there and uttered these things – because I am positive I heard them‖
(291a).32
He then proceeds to rule out all of his interlocutors by name. Rosamond Kent
Sprague hypothesizes that Socrates is here putting his own words in Clinias‘s mouth and
Crito‘s response that ―I certainly think it was some superior being‖33
(291a) is intended to
indicate that Crito knows this.34
Another possibility is that Plato is introducing an idea of
his own into a dialogue that cannot plausibly be attributed to any of the interlocutors.
Regardless of Plato‘s purposes in using this strange interjection into the dialogue, it
serves to highlight the passage for the reader, marking the passage as especially
important. Crito gives special praise to Clinias or whoever it was who made the
comparison, saying that whoever made such a comparison is in no need of education:
―Because, in my opinion, if he [Clinias] spoke like that, he needs no education, either
from Euthydemus or anyone else‖ (290e).35
This comparison of mathematics to hunting,
then, is something that Plato wants the reader to notice, and even goes out of his way to
use the intersection of the two narrative levels to highlight the comparison.
The distinction between productive and acquisitive crafts is further elaborated in
the Sophist. In the Sophist, the Eleatic Visitor divides all the crafts (―tōn ge technōn
pasōn‖) into two types (―eidē‖) (219a). The first type of craft he calls the ―productive‖
crafts, which closely accords with how Aristotle uses the term ―crafts‖ as a whole:
VISITOR: There‘s farming, or any sort of caring for any mortal
body; and there‘s also caring for things that are put together or
fabricated, which we call equipment; and there‘s imitation. The
right thing would be to call all those things by a single name.
THEAETETUS: How? What name?
31
{ΚΡ.} Σί ιέγεηο ζύ, ὦ ώθξαηεο; ἐθεῖλν ηὸ κεηξάθηνλ ηνηαῦη‘ ἐθζέγμαην; 32
ἀιι‘, ὦ δαηκόληε Κξίησλ, κή ηηο ηλ θξεηηηόλσλ παξὼλ αὐηὰ ἐθζέγμαην; ὅηη γὰξ ἢθνπζά γε ηαῦηα, εὖ
νἶδα. 33
ηλ θξεηηηόλσλ κέληνη ηηο ἐκνὶ δνθεῖ, θαὶ πνιύ γε. 34
Plato, Euthydemus, ed. Rosamond Kent Sprague (New York: Hackett, 1993), 36. 35
Μὰ Γί‘ νὐ κέληνη. νἶκαη γὰξ αὐηὸλ ἐγώ, εἰ ηαῦη‘ εἶπελ, νὔη‘ Δὐζπδήκνπ νὔηε ἄιινπ νὐδελὸο ἔη‘
ἀλζξώπνπ δεῖζζαη εἰο παηδείαλ.
19
VISITOR: When you bring anything into being that wasn‘t in
being before, we say you‘re a producer and that the thing you‘ve
brought into being is produced…Let‘s put them under the
heading of production (219a-b).36
The list of crafts here is in some ways quite striking, as Plato uses the language of
production, just as Aristotle will, but includes in production a number of activities that
don‘t seem productive. For instance, he includes caring for a mortal body and repairing
of equipment. In both of these cases, no new object is produced. Like Aristotle, he is
including ―accidental production‖, that is, the production some property in an already
existing object, such as the production of health in an ill body or of repairs in a broken
tool. The Visitor explicitly says that both caring for bodies and repairs ―bring something
into being that wasn‘t in being before‖, indicating that he does not intend to limit
production to the creation of new objects, but rather to any kind of production, including
accidental production.
The Visitor then discusses the acquisitive crafts. The passage is in many ways
quite similar to the passage in the Euthydemus, but there are some slight differences:
VISITOR: Next, consider the whole type that has to do with
learning, recognition, commerce, combat and hunting. None of
these creates anything. They take things that are or have come
into being, and they take possession of some of them with words
and actions, and they keep other things from being taken
possession of. For that reason it would be appropriate to call all
the parts of this type acquisition. (219c)37
In this passage, the Visitor repeats the distinction made previously in the Euthydemus.
The type of Aristotelian theoretical knowledge that has been included here has extended
beyond merely knowledge. In the Sophist, the Visitor includes all ―learning‖ (“to
mathēmatikon‖) as acquisitive which would include a broad range of sciences. There are
two other additions to this definition that were not in the Euthydemus. First, the Visitor
36
{ΞΔ.} Γεσξγία κὲλ θαὶ ὅζε πεξὶ ηὸ ζλεηὸλ πᾶλ ζκα ζεξαπεία, ηό ηε αὖ πεξὶ ηὸ ζύλζεηνλ θαὶ πιαζηόλ,
ὃ δὴ ζθεῦνο ὠλνκάθακελ, ἣ ηε κηκεηηθή, ζύκπαληα ηαῦηα δηθαηόηαη‘ἂλ ἑλὶ πξνζαγνξεύνηη‘ ἂλ ὀλόκαηη.
{ΘΔΑΙ.} Πο θαὶ ηίλη;
{ΞΔ.} Πᾶλ ὅπεξ ἂλ κὴ πξόηεξόλ ηηο ὂλ ὕζηεξνλ εἰο νὐζίαλ ἄγῃ, ηὸλ κὲλ ἄγνληα πνηεῖλ, ηὸ δὲ ἀγόκελνλ
πνηεῖζζαί πνύ θακελ… Πνηεηηθὴλ ηνίλπλ αὐηὰ ζπγθεθαιαησζάκελνη πξνζείπσκελ. 37
{ΞΔ.} Σὸ δὴ καζεκαηηθὸλ αὖ κεηὰ ηνῦην εἶδνο ὅινλ θαὶ ηὸ ηῆο γλσξίζεσο ηό ηε ρξεκαηηζηηθὸλ θαὶ
ἀγσληζηηθὸλ θαὶ ζεξεπηηθόλ, ἐπεηδὴ δεκηνπξγεῖ κὲλ νὐδὲλ ηνύησλ, ηὰ δὲ ὄληα θαὶ γεγνλόηα ηὰ κὲλ
ρεηξνῦηαη ιόγνηο θαὶ πξάμεζη, ηὰ δὲ ηνῖο ρεηξνπκέλνηο νὐθ ἐπηηξέπεη, κάιηζη‘ ἄλ πνπ δηὰ ηαῦηα ζπλάπαληα
ηὰ κέξε ηέρλε ηηο θηεηηθὴ ιερζεῖζα ἂλ δηαπξέςεηελ.
20
mentions ―taking possession‖ of the object sought, while in the Euthydemus, the analogy
between hunting and mathematics only extendeds as far as ―discovery‖. The ―taking
possession‖ of the learning crafts here is said to take place with ―words‖ (―tois logois”).
The second curious element is that Plato includes ―keeping other things from
being taken possession of‖ as a part of the acquisitive crafts. It is not immediately
obvious that keeping something safe ought to be considered an acquisitive craft; after all,
one does not actually acquire anything by keeping it safe. Republic I may provide some
clue as to why he considers guarding to be an acquisitive craft. In that dialogue, Socrates
points out that the person who is best at guarding something is also the best at stealing
(that is, acquiring) something:
And the one who is best guardian of an army is the very one who
can steal the enemy‘s plans and dispositions?
Certainly.
Whenever someone is a clever guardian, then, he is also a clever
thief.
Probably so.
If a just person is clever at guarding money, therefore, he must
also be clever at stealing it. (334a)38
In the Republic, Socrates claims that guarding and stealing are somehow the same craft.39
In summary, Plato provides a division of all crafts into the productive and the
acquisitive. Not only does Plato not use the Aristotelian definition of ―technē‖, but he
explicitly denies it. The productive crafts include all production of artifacts, but also all
therapeutic crafts, including medicine, as they produce properties in objects. The
acquisitive crafts include all hunting crafts, protection crafts and also intellectual crafts.
This division of all crafts will be important in ascertaining their relationships.
38
Ἀιιὰ κὴλ ζηξαηνπέδνπ γε ὁ αὐηὸο θύιαμ ἀγαζόο, ὅζπεξ θαὶ ηὰ ηλ πνιεκίσλ θιέςαη θαὶ βνπιεύκαηα
θαὶ ηὰο ἄιιαο πξάμεηο;
Πάλπ γε.
Ὅηνπ ηηο ἄξα δεηλὸο θύιαμ, ηνύηνπ θαὶ θὼξ δεηλόο.
Ἔνηθελ.
Δἰ ἄξα ὁ δίθαηνο ἀξγύξηνλ δεηλὸο θπιάηηεηλ, θαὶ θιέπηεηλ δεηλόο. 39
I will discuss in detail the senses in which this might be true in Chapter Two.
21
5 Crafts as Knowledge
While Plato uses the term ―technē‖ in a very broad, conventional way, a common
claim reoccurs in several dialogues: crafts are knowledge. That crafts are considered to
be knowledge will be very important to understanding the relationships between the
crafts discussed in other dialogues, as well as their relationships to the meta-crafts of
measurement and dialectic. In fact, in many dialogues, the term ―technē‖ or ―craft‖ and
the term ―epistēmē‖ or ―knowledge‖ are actually used interchangeably. In this section, I
will first show how Plato on numerous occasions throughout his corpus uses the terms
―technē‖ and ―epistēmē‖ synonomously, and then will focus on the discussion of the
teachability of crafts in the Protagoras, in which both Socrates and Protagoras agree that
virtue can only be a craft if it is teachable.
5.1 The Co-Extensivity of “Technē” and “Epistēmē”
(Principle #2: Crafts and Knowledge are co-extensive.)
On more than one occasion, Plato slips back and forth between the use of
―technē‖ (craft) and ―epistēmē‖ (knowledge) as though they were synonymous. In the
Charmides, Socrates uses the terms ―technē‖ and ―epistēmē‖ virtually interchangeably.40
At some points, he calls what one might expect to call a craft ―epistēmē‖ instead. For
instance, he refers to medicine as a kind of knowledge: ―‗Then medicine, too,‘ I said, ‗is
knowledge of health‘‖ (165c).41
Similarly, a few lines later, he calls housebuilding a kind
of knowledge as well: ―And if you ask me about housebuilding, which is knowledge of
building houses…‖ (165d).42
As medicine and housebuilding are paradigmatic examples
of crafts, for Socrates to here refer to both of them as kinds of ―knowledge‖ is especially
interesting.43
The use for housebuilding is a curious construction. He calls
housebuilding ―knowledge of housebuilding‖ or “epistēmēn tou oikodomein‖, attaching
40
David Roochnik, who normally translates the term ―epistēmē‖ as ―science‖ finds the interchangeability
sufficiently awkward that he chooses to leave ―epistēmē” untranslated when discussing the Charmides
(David Roochnik, "Socrates' Use of the Techne-Analogy", Journal of the History of Philosophy, 24 [1986]:
298). 41
Οὐθνῦλ θαὶ ἰαηξηθή, ἔθελ, ἐπηζηήκε ἐζηὶλ ηνῦ ὑγηεηλνῦ; 42
Καὶ εἰ ηνίλπλ κε ἔξνην ηὴλ νἰθνδνκηθήλ, ἐπηζηήκελ νὖζαλ ηνῦ νἰθνδνκεῖλ. 43
In fact, the use of ―technē‖ to describe housebuilding is so traditional that the term ―technē” actually has
its etymological roots in the word ―tekton‖ or carpenter.
22
an infinitive to knowledge. This indicates that housebuilding may somehow be or be of
something active (it is, after all, an infinitive of a verb), but at the same time, it ought to
be considered a form of knowledge. The dialogue of the Charmides slips back and forth
between the words “technē‖ and ―epistēmē‖ as though they had them same meaning.44
This same interchangeability reappears in the Statesman. In that dialogue, the
Visitor purports to be dividing all knowledge (―epistēmē‖): ―…we must make our minds
think of all sorts of knowledge there are as falling in two classes‖ (258c). However, a
little while later, when dividing knowledge, he divides all knowledge exhaustively into
two sorts of craft:
VISITOR: Well then: isn‘t it the case that arithmetic and some of
the other crafts that are akin to it don‘t involve any practical
knowledge?‖
YOUNG SOCRATES: That‘s so.
VISITOR: Whereas for their part, carpentry and manufacture as a
whole have their knowledge (―epistēmēn‖) as it were naturally
bound up with practical actions… (258d-e)45,46
Notice how the Eleatic Visitor recharacterizes the division here. He had at first said that
he was dividing all types of knowledge (―epistēmē‖), but when he actually divided that
knowledge, he divides into two sorts of crafts: first, arithmetic and crafts akin to it, and
then into carpentry (“tekntonikēn‖) and other sorts of manufacture. It appears, then, that
there is no knowledge that is not itself a craft.47
Moreover, a page later, the Visitor uses
the same sort of interchangeability between crafts and knowledge that Socrates made use
44
John Lyons discussed this interchangeability in detail in his book Structural Semantics: an analysis of
part of the vocabulary of Plato. He concluded that the phrases ―Socrates is a housebuilder‖ (―Ʃσθξάηεο
εἶλαη νἰθνδόκνο‖), ―Socrates has the housebuilding craft‖ (―σθξάηεο ἔρεηλ ηὴλ νἰθνδνκηθὴλ ηερλήλ‖),
―Socrates knows housebuilding things‖ (―σθξάηεο ἐπίζηαζζαη νἰθνδόκηθα‖), and ―Socrates knows [how]
to build houses‖ (―σθξάηεο ἐπίζηαζζαη ηὸ νἰθνδνκεῖλ‖) were transformable in Plato, and that Plato would
regularly switch between them (John Lyons, Structural Semantics: An Analysis of Part of the Vocabulary
of Plato [Oxford: Basil Blackwell, 1963], 173). 45
{ΞΔ.} Ἆξ‘ νὖλ νὐθ ἀξηζκεηηθὴ κὲλ θαί ηηλεο ἕηεξαη ηαύηῃ ζπγγελεῖο ηέρλαη ςηιαὶ ηλ πξάμεώλ εἰζη, ηὸ
δὲ γλλαη παξέζρνλην κόλνλ;
{ΝΔ. Ω.} Ἔζηηλ νὕησο.
{ΞΔ.} Αἱ δέ γε πεξὶ ηεθηνληθὴλ αὖ θαὶ ζύκπαζαλ ρεηξνπξγίαλ ὥζπεξ ἐλ ηαῖο πξάμεζηλ ἐλνῦζαλ ζύκθπηνλ
ηὴλ ἐπηζηήκελ θέθηεληαη, θαὶ ζπλαπνηεινῦζη ηὰ γηγλόκελα ὑπ‘αὐηλ ζώκαηα πξόηεξνλ νὐθ ὄληα. 46
I have deviated from Rowe‘s translation here, as he has Socrates explicitly say that carpentry and
manufacture are crafts, while the Greek is more ambiguous. 47
That carpentry and manufacture are ―epistemē‖ but not ―technē‖ is an interpretation compatible with this
passage. However, such a category would be odd, as a category of non-craft, manual epistēmē would be
otherwise unprecedented.
23
of in the Charmides, when he characterizes his earlier claim as being one of division of
technai rather than epistēmai: ―And I suppose it [calculation] belongs absolutely to the
theoretical sorts of craft‖ (259e).48
However, earlier it was knowledge that he was
dividing into theoretical and practical, not crafts; instead, he now says that what he had
been doing is dividing crafts. The Visitor here is using the terms ―technē‖ and
―epistēmē‖ interchangeably, just as in the Charmides. Even in the late dialogues, then,
Plato continued to treat ―technē‖ and ―epistēmē‖ largely interchangeably.
5.2 The Teachability of the Crafts
Principle #3: Crafts are teachable
In the Protagoras, Socrates asks Protagoras what it is that he claims to teach.
After Protagoras gives a somewhat longwinded explanation, Socrates summarises what
Protagoras has just claimed: ―You appear to be talking about the art of citizenship, and to
be promising to make men good citizens.‖ (319a). Protagoras agrees that this is exactly
what he is claiming. However, Socrates says that he has doubts about whether or not
virtue is teachable. In his argument that virtue is not teachable, Socrates explicitly
contrasts what he calls ―technical‖ skills with good citizenship:
And I observe that when we convene in the Assembly and the
city has to take some action on a building project, we send for
builders to advise us…and so forth for everything that is
learnable and teachable (319b)49
However, he argues that non-crafts including citizenship are not teachable:
But when it is a matter of deliberating on city management,
anyone can stand up and advise them…The reason for this is
clear: They do not think that this can be taught. (319d-e) 50
48
{ΞΔ.} Σλ γλσζηηθλ γε νἶκαη παληάπαζη ηερλλ. 49
ὁξ νὖλ, ὅηαλ ζπιιεγκελ εἰο ηὴλ ἐθθιεζίαλ, ἐπεηδὰλ κὲλ πεξὶ νἰθνδνκίαο ηη δέῃ πξᾶμαη ηὴλ πόιηλ, ηνὺο
νἰθνδόκνπο κεηαπεκπνκέλνπο ζπκβνύινπο πεξὶ ηλ νἰθνδνκεκάησλ, ὅηαλ δὲ πεξὶ λαππεγίαο, ηνὺο
λαππεγνύο, θαὶ ηἆιια πάληα νὕησο, ὅζα γνῦληαη καζεηά ηε θαὶ δηδαθηὰ εἶλαη. 50
ἐπεηδὰλ δέ ηη πεξὶ ηλ ηῆο πόιεσο δηνηθήζεσο δέῃ βνπιεύζαζζαη, ζπκβνπιεύεη αὐηνῖο ἀληζηάκελνο πεξὶ
ηνύησλ ὁκνίσο κὲλ ηέθησλ, ὁκνίσο δὲ ραιθεὺο ζθπηνηόκνο, ἔκπνξνο λαύθιεξνο, πινύζηνο πέλεο, γελλαῖνο
ἀγελλήο, θαὶ ηνύηνηο νὐδεὶο ηνῦην ἐπηπιήηηεη ὥζπεξ ηνῖο πξόηεξνλ, ὅηη νὐδακόζελ καζώλ, νὐδὲ ὄληνο
δηδαζθάινπ νὐδελὸο αὐηῶ, ἔπεηηα ζπκβνπιεύεηλ ἐπηρεηξεῖ· δῆινλ γὰξ ὅηη νὐρ γνῦληαη δηδαθηὸλ εἶλαη.
24
Socrates‘ argument depends on the claim that all crafts are teachable51
. The first list is of
the crafts, and he claims that people believe that crafts can be taught, which is why they
listen to experts. However, good citizenship cannot be taught, and therefore is not a craft
and does not admit of specialization. As a result, the assembly allows anyone to speak on
matters of good citizenship.
When Protagoras defends his claim to be able to teach virtue, he at no point
denies that crafts can be taught (in fact, he cannot deny this without putting himself out of
business, as he claims to be a teacher of a craft). Instead, he argues that good citizenship
is taught, it is just that it is a craft taught universally, much as grammar is taught
universally. In his myth of the distribution of the crafts by Zeus, Protagoras claims that
Zeus distributed the craft of virtue to everyone, rather than to just a few:
Hermes asked Zeus how he should distribute shame and justice
to humans.52
‘Should I distribute them [shame and justice] as the
other crafts were?‘…‘To all,‘ said Zeus, ‗and let all have a
share.‘ (322c-d)53
As a result, the crafts of shame and justice (which Protagoras is equating with the craft of
citizenship) are distributed to all, but that does not imply that they are not teachable.
Rather, they are crafts that everyone is taught, right from an early age, and the whole
society is constantly teaching virtue to one another, especially to children:54
When so much care and attention is paid to virtue, Socrates, both
in public and private, are you still puzzled about virtue being
teachable? The wonder would be if it were not teachable‖
(326e) 55
51
Note that this is not exactly the same issue as whether or not crafts are learnable. John Lyons notes, in
Plato‘s Greek, ―learning‖ has the same consequential relationship to ―knowing‖ as ―becoming‖ has to
―being‖ (―καλζάλεηλ‖:―ἐπίζηαζζαη‖::―γίγλεζζαη‖:―εἶλαη‖), and therefore, it is analytic that becoming a
craftsperson requires learning a craft. However, learning does not necessarily imply teaching. One could,
for example, learn from experience or through calculation (John Lyons, Structural Semantics: An Analysis
of Part of the Vocabulary of Plato [Oxford: Basil Blackwell, 1963], 155). 52
It is counter-intuitive to call shame and justice crafts, and there isn‘t space to justify Protagoras‘ use here.
It is tied to Protagoras‘s claim that beliefs about what are just and shameful are both themselves taught. 53
ἐξσηᾷ νὖλ ξκῆο Γία ηίλα νὖλ ηξόπνλ δνίε δίθελ θαὶ αἰδ ἀλζξώπνηο· ―Πόηεξνλ ὡο αἱ ηέρλαη
λελέκεληαη, νὕησ θαὶ ηαύηαο λείκσ; λελέκεληαη δὲ ὧδε· εἷο ἔρσλ ἰαηξηθὴλ πνιινῖο ἱθαλὸο ἰδηώηαηο, θαὶ νἱ
ἄιινη δεκηνπξγνί· θαὶ δίθελ δὴ θαὶ αἰδ νὕησ ζ ἐλ ηνῖο ἀλζξώπνηο, ἠ ἐπὶ πάληαο λείκσ;‖ ―πὶ πάληαο,‖
ἔθε ὁ Εεύο, ―θαὶ πάληεο κεηερόλησλ.‖ 54
Protagoras slips back-and-forth between justice and virtue a few times in this speech. 55
ηνζαύηεο νὖλ ηῆο ἐπηκειείαο νὔζεο πεξὶ ἀξεηῆο ἰδίᾳ θαὶ δεκνζίᾳ, ζαπκάδεηο, ὦ ώθξαηεο, θαὶ ἀπνξεῖο εἰ
δηδαθηόλ ἐζηηλ ἀξεηή; ἀιι‘ νὐ ρξὴ ζαπκάδεηλ, ἀιιὰ πνιὺ κᾶιινλ εἰ κὴ δηδαθηόλ.
25
Protagoras disagrees with Socrates that something‘s being teachable implies that it is
specialized. However, he agrees with Socrates that all crafts are teachable. While the
Protagoras is an extremely complex dialogue ending in aporia (confusion) and in which
it is not clear that Socrates leaves the victor, that both interlocutors in a dialogue agree on
a shared claim, in this case that crafts are teachable, provides strong evidence that it is a
view that Plato thinks has rational support
In this section, I have provided both several examples of near-co-extensive use of
―technē‖ and ―epistēmē‖ by Plato and also an explicit argument that crafts must be
teachable because they are knowledge in the Protagoras. That crafts are a form of
knowledge is important to understanding their relationship with each other and with
meta-crafts (those crafts that are included in all crafts) that I will dicuss in the next
section.
6 Crafts and Dialectic
Principle #4: Crafts are dependent on dialectic for a determinate subject matter and
to systematically approach causal relationships.
Principle #4a: Crafts have a determinate subject matter
Principle #4b: Crafts depend on the understanding of causal relationships.
The crafts have a special relationship with two meta-crafts, dialectic and
measurement. These meta-crafts are required for any crafts to function. The relationship
with dialectic is introduced in the Gorgias and discussed in detail in the Phaedrus. By
dialectic, I am referring to the craft of collecting things into unified kinds and then
dividing those unified kinds according to appropriate divisions. This relationship
between other crafts and dialectic is especially important to medicine, as the relationship
described in the Phaedrus putatively derives from medicine and is in fact called ―the
method of Hippocrates‖. In the Phaedrus, Socrates asserts that all productive crafts
depend on dialectic in order to properly function. They are dependent on dialectic for
two reasons. First, through collection, dialectic provides the crafts with an intelligible
object; without it, they would not be able to differentiate between the possible ends of a
craft. Second, dialectic provides appropriate divisions within its subject matter, allowing
26
the producer to discover the causal relationships between its parts. Dialectic then serves
as the first of two of what I shall call ―meta-crafts‖, crafts that, according to Plato, any
craft must use in order to be considered properly a craft at all.
The Gorgias, like the Protagoras, includes a very important passage in which the
craft-status of a particular practice is challenged. Since Plato uses the term ―technē‖ in
such a broad, conventional way, passages where something‘s status as a craft is directly
challenged are especially important. These challenges allow us to determine positive
criteria for a craft by taking note of the bases on which characters deny that something is
a craft. In the Gorgias, Socrates explicitly excludes four practices from being considered
crafts: rhetoric, sophistry, pastry-baking and cosmetics. As he argues directly against the
craft status of pastry-baking and cosmetics for his argument, I will focus on these latter
two. He criticizes pastry-baking and cosmetics for two reasons, each of which appears to
be sufficient to exclude them as crafts: they aim at pleasure rather than the good and they
lack an understanding of their objects.56
First, he argues that pastry-baking and cosmetics are kinds of flattery (―kolakeia‖)
that do not have any knowledge, but only guess as what is most pleasant, rather than
knowing what is best:
Now flattery takes notice of them, and – I won‘t say by knowing,
but only by guessing – divides itself into four, masks itself in
each of the parts, and then pretends to be the characters of the
masks. It takes no thought at all of whatever is best; with the
lure of what‘s most pleasant at the moment, it sniffs out folly and
hoodwinks it, so that it gives the impression of being most
deserving. (464c-d)57
The contrast here is two-fold: on the one hand, pastry-baking is flattery while medicine is
a craft, because pastry-baking tries to produce pleasure, while medicine tries to produce
56
As Raphael Woolf notes, not aiming at a good and lacking an account are not simply two independent
necessary conditions for something‘s being a craft as opposed to a practice. They are connected. A
practice without a good cannot be systematic. His approach is different from mine, however, as he seeks
an explanation of the connection within the Gorgias itself, while I seek the explanation by using the
Phaedrus to tie the two criteria together (Raphael Woolf, "Why is Rhetoric Not a Skill?", History of
Philosophy Quarterly, 21 [2004]: 119-30). David Levy, on the other hand, treats the Socratic claim that
rhetoric is unsystematic as an empirical, historical claim about contemporary rhetors rather than a
theoretical claim (David Levy, "Technē and the Problem of Socratic Philosophy in the Gorgias", Apeiron,
38 [2005]: 205-8). 57
θνιαθεπηηθὴ αἰζζνκέλε—νὐ γλνῦζα ιέγσ ἀιιὰ ζηνραζακέλε—ηέηξαρα ἑαπηὴλ δηαλείκαζα, ὑπνδῦζα
ὑπὸ ἕθαζηνλ ηλ κνξίσλ, πξνζπνηεῖηαη εἶλαη ηνῦην ὅπεξ ὑπέδπ, θαὶ ηνῦ κὲλ βειηίζηνπ νὐδὲλ θξνληίδεη, ηῶ
δὲ ἀεὶ δίζηῳ ζεξεύεηαη ηὴλ ἄλνηαλ θαὶ ἐμαπαηᾷ, ὥζηε δνθεῖ πιείζηνπ ἀμία εἶλαη.
27
what is best; on the other hand, flattery guesses while medicine knows.58
The pleasure is
produced because pastry-baking pretends to be medicine, in the sense that it fools the
tongue into enjoyment that is properly reserved for food that is healthy. Cosmetics is
similar; it makes someone appear beautiful, while gymnastics makes him or her truly
beautiful. As such, Socrates excludes pastry-baking and cosmetics because they do not
have a unique subject matter. They are logically parasitic on medicine and gymnastics
and seek to imitate their objects, and produce pleasure in as much as they are able to
successfully manipulate the pleasure naturally associated with health and beauty.59
However, Socrates goes further than merely chastising pastry-baking and
cosmetics for being parasitic and imitative.60
He also says that they lack knowledge of
their objects, claiming that as a result, it is not a craft, but rather a kind of ―practice‖ or
―empeiria‖:61
And I say that it isn‘t a craft, but a practice, because it has no
account of the nature of whatever things it applies by which it
applies them, so that it‘s unable to state the cause of each thing.
(465a) 62
58
Though the quoted portion of the passage doesn‘t mention medicine, it is the craft being contrasted with
the kolakeia of pastry-baking. 59
My solution differs from that of Woolf in the following important respect. He allows that there might be
a systematic craft of producing pleasure, but that the problem with pastry-bakers and sophists is that they
pretend to be doctors and politicians to the hearers. The problem, then, isn‘t with their activity, but with
what they pretend their activity is (Raphael Woolf, "Why is Rhetoric Not a Skill?", History of Philosophy
Quarterly, 21 [2004]: 119-30). However, the problem with this interpretation is that it allows for the
possibility of a pure confectionary craftsman who clearly places health warnings on his or her tasty
products. Socrates intends to rule out even this, because the pleasure from food doesn‘t simply fool the
taster but the tongue. It is not only the pastry baker who is imitating a doctor, but the pastry that is
imitating food. The practice of pastry baking lacks any product of its own because its product is merely an
imitation of the product of another craft. 60
For an interesting account of in just what way pleasure is parasitic on the good state of the body, see Lee
Franklin, "Technē and Teleology in Plato‘s Gorgias", Apeiron, 38 (2005): 229-55. 61
Note that this is not the distinction between ―crafts‖ and ―practices‖ made by Alasdair MacIntyre. He
uses the term ―practices‖ to refer to social practices whose end is virtue: ―By a practice I am going to mean
any coherent and complex form of socially established co-operative human activity through which goods
internal to that form of activity are realised in the course of trying to achieve those standards of excellence
which are appropriate to, and partially definitive of, that form of activity, with the result that human powers
to achieve excellence, and human conceptions of the ends and goods involved, are systematically
extended‖ (Alasdair MacIntyre, After Virtue, 3rd ed. [South Bend: University of Notre Dame Press, 2007],
187). 62
ηέρλελ δὲ αὐηὴλ νὔ θεκη εἶλαη ἀιι‘ ἐκπεηξίαλ, ὅηη νὐθ ἔρεη ιόγνλ νὐδέλα ᾧ πξνζθέξεη ἃ πξνζθέξεη
ὁπνῖ‘ ἄηηα ηὴλ θύζηλ ἐζηίλ, ὥζηε ηὴλ αἰηίαλ ἑθάζηνπ κὴ ἔρεηλ εἰπεῖλ.
28
This lack of knowledge is two-fold and connected: first, it lacks an account of the nature
of what it applies to, and second, it is unable to state the causes of each thing.63
This
leaves the reader with two significant puzzles. First, how is this connected to his earlier
discussion of knowing what is best and seeking what is best? Socrates excludes the
possibility that someone might know the nature of something, but not seek what is best,
or alternatively, seek what is best but not know its nature. Somehow, seeking what is
best and knowing the nature of something are connected, yet we are not told why.
Second, how does knowing the nature of something relate to knowing the cause of each
thing?64
Given the ―so that‖ in the above quotation, the inability to state causes is
somehow a result of not knowing the nature of something, but it is not immediately
obvious that this should follow.
Fortunately, these puzzles are addressed in detail in the Phaedrus,65
when
Socrates describes what he calls the ―Method of Hippocrates‖. In that dialogue, Socrates
addresses the same question that is being addressed in the Gorgias passage quoted above:
is rhetoric a craft? In the Phaedrus, Socrates opens up the possibility that there is a kind
of rhetoric that might be considered a craft, but sets a series of strict conditions.
Importantly, these conditions are set as conditions for any craft, and fortunately for the
purposes of this dissertation, the example is a medical one. In order to be considered a
craft, a craft must follow the ―Method of Hippocrates‖, which depends on a method of
collection and division referred to as dialectic. In the rest of this section, I will describe
this method of collection and division and how Socrates argues that the crafts depend on
dialectic.
63
David Levy raises the interesting possibility that Socrates, at least in part, criticizing the method of
developing crafts that Polus provides at Gorgias 448c, in which Polus claimed that craft comes from
practice (―empereia‖) (David Levy, "Technē and the Problem of Socratic Philosophy in the Gorgias",
Apeiron, 38 [2005]: 201). If one translates the word ―heurēmenai‖ as ―discovered‖ rather than Zeyl‘s
―devised‖, the unsystematic nature of Polus‘ claim becomes apparent. 64
This question is also raised by Lee Franklin in ―Technē and Teleology in Plato‘s Gorgias‖. He too
argues that knowledge of causal (“aitia‖) relationships depends on knowing the essential property of
something: ―Generally, the regular behavior of a property – the full range of its law-like relationships –
results from what it is, i.e. its essence‖. Even if this is seems correct, we do not hear from Plato in the
Gorgias why this is the case (Lee Franklin, "Technē and Teleology in Plato‘s Gorgias", Apeiron, 38 [2005]:
238). 65
I will not take a position on which dialogue was written first. However, the Phaedrus does appear to
significantly flesh out some of the assumptions of the Gorgias passage. This could be a result of the
Phaedrus being written to clarify portions of the Gorgias, but it could equally be the result of the Gorgias
being written with the assumption that the reader was already familiar with the ―Hippocratic method‖
described in the Phaedrus.
29
In the Phaedrus, Socrates elaborates on the requirements presented briefly in the
Gorgias. First, he explains the connection between seeking the good in a craft and
knowing the nature of one‘s subject. In order to explain this connection, he provides the
example of a physician who manipulates the body without reference to its good, using the
comic example of someone who is able to induce vomiting or bowel movements to no
end:
SOCRATES: Suppose someone came to your friend Eryximachus
or his father Acumenus and said: ‗I know treatments to raise or
lower (whichever I prefer) the temperature of people‘s bodies; if
I decide to, I can make them vomit or make their bowels move,
and all sorts of things. On the basis of this knowledge, I claim to
be a physician; and I claim to be able to make others physicians
as well by imparting it to them.‘ What do you think they would
say when they heard that?
PHAEDRUS: What could they say? They would ask him if he
also knew to whom he should apply such treatments, when and
to what extent. (268a-c)66
Socrates‘ point here is a straightforward and obvious one: the ability to manipulate the
body in various ways is not medicine; one needs to do so with some sort of purpose in
mind. Socrates does have a place for the sort of knowledge that the mere technician has.
He calls these, instead, the ―preliminaries‖ of medicine. That is, they are techniques that
one learns prior to learning medicine proper, and include the ability to manipulate the
body in various ways. Medicine itself allows the physician to manipulate the body in the
right way.
It is this connection between preliminaries and medicine proper that provide the
connection missing in the Gorgias between knowing what is best and knowing the nature
of one‘s object. In the Gorgias, Socrates claims that pastry-baking and cosmetics are not
crafts on the grounds of not caring for the best and not understanding the nature of their
object. The Phaedrus makes this connection explicit. Here, Socrates argues that
dialectic allows the craftsperson to use appropriate collection and division to discover the
66
{Ω.} Δἰπὲ δή κνη· εἴ ηηο πξνζειζὼλ ηῶ ἑηαίξῳ ζνπ ξπμηκάρῳ ἠ ηῶ παηξὶ αὐηνῦ Ἀθνπκελῶ εἴπνη ὅηη
―γὼ ἐπίζηακαη ηνηαῦη‘ ἄηηα ζώκαζη πξνζθέξεηλ, ὥζηε ζεξκαίλεηλ η‘ ἐὰλ βνύισκαη θαὶ ςύρεηλ, θαὶ ἐὰλ
κὲλ δόμῃ κνη, ἐκεῖλ πνηεῖλ, ἐὰλ δ‘ αὖ, θάησ δηαρσξεῖλ, θαὶ ἄιια πάκπνιια ηνηαῦηα· θαὶ ἐπηζηάκελνο αὐηὰ
ἀμη ἰαηξηθὸο εἶλαη θαὶ ἄιινλ πνηεῖλ ᾧ ἂλ ηὴλ ηνύησλ ἐπηζηήκελ παξαδ,‖ ηί ἂλ νἴεη ἀθνύζαληαο εἰπεῖλ;
{ΦΑΙ.} Σί δ‘ ἄιιν γε ἠ ἐξέζζαη εἰ πξνζεπίζηαηαη θαὶ νὕζηηλαο δεῖ θαὶ ὁπόηε ἕθαζηα ηνύησλ πνηεῖλ, θαὶ
κέρξη ὁπόζνπ;
30
nature of the whole subject of his or her own craft. When discussing rhetoric, which he
describes as ―psychagōgē‖ or ―soul-leading‖, Socrates argues that one cannot understand
the subject unless one understands the nature of the whole soul: ―Do you think then that it
is possible to reach a serious understanding of the nature of the soul without
understanding the nature of the whole?‖ (270b).67
What is meant by ―whole‖ in the above passage is contentious, and there are
various interpretations.68
Paul Woodruff interprets it to mean, ―the whole universe‖ and,
in fact, translates this phrase ―…the nature of the world as a whole?‖, despite there being
no Greek to correlate to the phrase ―the world as.‖69
Jacques Jouanna translates it in a
similar way: ―...the nature of the universe.‖70
This interpretation is incorrect, as it is not
consistent with other portions of the text, which indicate that Socrates is referring to the
product or subject of the craft, considered as a whole. Socrates considers the case of
what I shall call the ―pulp fiction‖ author, who is analogous to the mere technician in
medicine: he or she knows how to manipulate emotions, but does not know how to put
those emotional manipulations together into a whole work of tragedy:
PHAEDRUS: Oh, I am sure they too would laugh at anyone who
thought a tragedy was anything other than the proper
arrangement of these things: They have to fit with one another
and with the whole. (268d)71
Phaedrus here is referring to the arrangement of the parts in the whole work of tragedy,
not the whole universe. The word ―whole‖, then, has already been used just a page
earlier, and what is meant by it is the product of the craft, considered as a whole, not the
whole world.72
67
{Ω.} Ψπρῆο νὖλ θύζηλ ἀμίσο ιόγνπ θαηαλνῆζαη νἴεη δπλαηὸλ εἶλαη ἄλεπ ηῆο ηνῦ ὅινπ θύζεσο; 68
G.E.R. Lloyd notes that this dispute goes back to the ancient world, and he finds the dispute so
intractable that he seeks separate textual antecedents in the Hippocratic corpus for each interpretation. He
himself takes no position (G. E. R. Lloyd, Methods and Problems in Greek Science [Cambridge:
Cambridge University Press, 1991], 202-03). 69
Paul Woodruff. Plato, Plato: Complete Works, ed. John Cooper (Indianapolis: Hackett, 1997), 547. 70
Jacques Jouanna, Hippocrates (Baltimore: Johns Hopkins University Press, 1999), 6. Much of the debate
about to which text Plato is referring assumes that the Phaedrus ought to be interpreted in this way. If my
interpretation is correct, it would open possible alternative sources for the ―Hippocratic method‖ in the
Phaedrus. 71
{ΦΑΙ.} Καὶ νὗηνη ἄλ, ὦ ώθξαηεο, νἶκαη θαηαγειῶελ εἴ ηηο νἴεηαη ηξαγῳδίαλ ἄιιν ηη εἶλαη ἠ ηὴλ ηνύησλ
ζύζηαζηλ πξέπνπζαλ ἀιιήινηο ηε θαὶ ηῶ ὅιῳ ζπληζηακέλελ. 72
Woodruff and Jouanna‘s translations were perhaps influenced by a peculiar claim made by Socrates a
few interchanges earlier, that all crafts require ethereal speculation about nature: ―All the great arts require
endless talk and ethereal speculation about nature‖ (270a) . However, Socrates does not mean the whole of
31
One needs to understand that subject or product as a unity, or else the craft has no
product and is reduced to a series of techniques, like the mere medical technician or the
pulp-fiction playwright. In this way, the introduction of wholes and natures connects the
two parts of Socrates‘ charge against pastry-baking and cosmetics in the Gorgias. The
pastry-baker is unable to consider the good, because the good of medicine is the body,
considered as a whole. Since the pastry-baker does not understand the nature of the
body, he or she is forced instead to simply appeal to its surrounding pleasures, unable to
consider the body‘s good because he or she is unable to consider the body‘s nature.
Understanding of the nature of the subject matter is what allows the crafts to use their
techniques in tandem so as to produce a whole product. This connects the two concerns
of the Gorgias, by requiring understanding of the functioning of the body as a whole as a
condition of attending to its good.
Dialectic is that craft that allows a craftsperson to understand this whole and the
relationship between its parts to each other and to outside causes.73
The method of
dialectic described in the Phaedrus is the method of collection and division. Dialectic is
what Socrates calls a ―systematic art‖, and it enables one to perform the two tasks of
collection and division: ―…there were in it [love] two kinds of things the nature of which
it would be quite wonderful to grasp by means of a systematic art‖ (265c-d).74
The first
nature, as he immediately tells us, but rather, the nature of the subject of the craft. Pericles had studied
with Anaxagoras, who had said a great deal about the nature of the soul, the proper subject of rhetoric (a
kind of soul-leading):
―Anaxagoras…understood the nature of mind and mindlessness – just
the subject on which Anaxagoras had the most to say. From this, I
think, he [Pericles] drew for the art of rhetoric what was most useful to
it‖ (270a)
The whole that is understood in learning a craft is the whole subject or product of the craft. The playwright
understands the whole product of his or her work, while the rhetorician understands the nature of the soul,
the subject of his or her craft. 73
Franklin holds that there are actually two different levels of causal relationships, teleological and material
relationships: ―Within certain limitations, the craftsman must understand both levels, teleological and
material, and the way they interact. Indeed, although the craftsperson‘ primary concern is the law-like
relations in the teleological system of the technē, the work of the technē is accomplished by harnessing the
non-teleological law-like relations of the underlying phenomena‖ (Lee Franklin, "Technē and Teleology in
Plato‘s Gorgias", Apeiron, 38 [2005]: 253). However, though both the Gorgias and the Phaedrus make a
distinction between technique and proper application, neither says anything about two distinct, law-like
bodies of causal knowledge. On the contrary, both the reference to the pastry-baker as ―guessing‖ and the
reference to the vomit-inducer as having a ―few‖ potions imply that there is not law-like understanding at
all without understanding of the good of the whole subject. 74
ηνύησλ δέ ηηλσλ ἐθ ηύρεο ῥεζέλησλ δπνῖλ εἰδνῖλ, εἰ αὐηνῖλ ηὴλ δύλακηλ ηέρλῃ ιαβεῖλ δύλαηηό ηηο, νὐθ
ἄραξη.
32
part of dialectic, collection, is what enables one to understand wholes. Socrates explains
this near the beginning of the discussion of wholes:
SOCRATES: The first consists in seeing together things that are
scattered about everywhere and collecting them into one kind, so
that by defining each thing we can make clear the subject of any
instruction we wish to give. (265d)75
.
Without collection into kinds and therefore dialectic, it would be impossible to appreciate
one‘s subject as a whole and therefore to have a definite product. The single subject and
its good are what give these techniques their unity.
The second part of dialectic is equally important to the development of all crafts,
as it is what enables the craftsperson to understand the causal relationships necessary for
production in a craft. This second part of dialectic is division, which Plato likens to
cutting up the world like a good butcher: ―This, in turn, is to be able to cut up each kind
according to its species along its natural joints and to try not to splinter any part, as a bad
butcher might do‖ (265e).76
Only by correctly ―cutting up‖ the parts of the subject of the
craft, can the craftsperson systematically understand the causal relationships between
those parts. In the Phaedrus, Socrates is more explicit about the kind of causes that a
craftsperson must know. Having used dialectic to divide the subject into parts, the
craftsperson then examines the relationships between the parts.
The passage of the Phaedrus discussing the utility of division to crafts is complex,
so I will quote it in its entirety:
First, we must consider whether the object regarding which we
intend to become experts and capable of transmitting our craft is
simple or complex. Then, if it is simple, we must investigate its
power: What things does it have what power of acting upon? By
what things does it have what natural disposition to be acting
upon? If, on the other hand, it takes many forms, we must
enumerate them all and, as we did in the simple case, investigate
how each is naturally able to act upon what and how it has a
natural disposition to be acted upon by what. (270d-e)77
75
{Ω.} Δἰο κίαλ ηε ἰδέαλ ζπλνξληα ἄγεηλ ηὰ πνιιαρῇ δηεζπαξκέλα, ἵλα ἕθαζηνλ ὁξηδόκελνο δῆινλ πνηῇ
πεξὶ νὗ ἂλ ἀεὶ δηδάζθεηλ ἐζέιῃ. 76
{Ω.} Σὸ πάιηλ θαη‘ εἴδε δύλαζζαη δηαηέκλεηλ θαη‘ ἄξζξα ᾗ πέθπθελ, θαὶ κὴ ἐπηρεηξεῖλ θαηαγλύλαη
κέξνο κεδέλ, θαθνῦ καγείξνπ ηξόπῳ ρξώκελνλ. 77
πξηνλ κέλ, ἁπινῦλ ἠ πνιπεηδέο ἐζηηλ νὗ πέξη βνπιεζόκεζα εἶλαη αὐηνὶ ηερληθνὶ θαὶ ἄιινλ.
33
Using collection, one is able to understand the nature of the object, but that is not
sufficient for understanding a craft. In addition to understanding the nature of the object,
productive crafts require the understanding of causal relationships in order that the
desired goal can be produced. The Hippocratic method has two steps. First, one must
systematically divide the subject into parts if it is complex. Then, one must examine the
causal interaction of those parts with the parts of everything else. I will discuss each of
these parts of division‘s relationship with subjects in turn.
An appropriate division of the subject of a craft into parts is required in order that
the understanding of causal relationships be systematic. When describing the would-be
physician earlier, who knew how to make a person throw up or have a bowel movement
and thought that made him a physician, Phaedrus criticizes him for not understanding the
whole, but afterwards, Phaedrus adds a second criticism, that this physician‘s knowledge
is incomplete: ―I think they‘d say the man‘s mad if he thinks he‘s a doctor because he
read a book or happened to come across a few potions; he knows nothing of the craft‖
(268c).78
The putative physician who does not understand dialectic not only lacks
understanding of when to use those techniques he has, but he lacks complete techniques,
consisting only of a ―few potions‖. Division is what allows craft knowledge not to be
simply a ―few potions‖. With appropriate division of its subject into parts, medicine is
able to be systematic. It is not enough to simply say that medicine should understand
causal powers; without understanding how to properly divide the body into parts, one
would not even know what relationships one is looking for. Dialectic allows the
physician to understand the separate parts of the body in such a way that the search for
the causal relations that might benefit the body can be begun systematically.
Despite using medicine for his initial analogy, the extended discussion which
Plato provides is not about medicine, but rhetoric. He seeks to divide souls into different
types (genē), and then divide speech into different types (genē), and see what the causal
relationships are between each of them. Once all of these relationships have been noted,
one has understanding of the causal relationships required for understanding rhetoric.
δπλαηνὶ πνηεῖλ, ἔπεηηα δέ, ἂλ κὲλ ἁπινῦλ ᾖ, ζθνπεῖλ ηὴλ δύλακηλ αὐηνῦ, ηίλα πξὸο ηί πέθπθελ εἰο ηὸ δξᾶλ
ἔρνλ ἠ ηίλα εἰο ηὸ παζεῖλ ὑπὸ ηνῦ, ἐὰλ δὲ πιείσ εἴδε ἔρῃ, ηαῦηα ἀξηζκεζάκελνλ, ὅπεξ ἐθ‘ ἑλόο, ηνῦη‘ ἰδεῖλ
ἐθ‘ ἑθάζηνπ, ηῶ ηί πνηεῖλ αὐηὸ πέθπθελ ἠ ηῶ ηί παζεῖλ ὑπὸ ηνῦ; 78
{ΦΑΙ.} Δἰπεῖλ ἂλ νἶκαη ὅηη καίλεηαη ἅλζξσπνο, θαὶ ἐθ βηβιίνπ πνζὲλ ἀθνύζαο ἠ πεξηηπρὼλ θαξκαθίνηο
ἰαηξὸο νἴεηαη γεγνλέλαη, νὐδὲλ ἐπαΐσλ ηῆο ηέρλεο.
34
This could become quite an involved process, as the total number of interactions to be
studied would be the product of the number of types of souls and the number of types of
speech, and Socrates does not provide a detailed analysis of these relationships here.
However, dialectic is necessary to this method in order to even start the process: without
an appropriate division into types, it would be impossible to know which interactions to
look for.79
One puzzle in this section, however, is that Plato is using the term ―genē‖ in two
different ways: as types and as parts. When discussing the types of souls, he talks about
the different types of souls and the different types of speech. Each then has a certain kind
of result: for instance, insults to a vain person generally create the same response.
However, when discussing medicine, the division described is not a division into kinds
but a division into parts. Understanding how the body works is not simply a matter of
classifying people and determining what effects things have on different kinds of bodies,
but of actually understanding the parts of the body and how each of them is affected by
various treatments. In fact, his metaphor of ―cutting up nature according to its natural
joints‖ is clearly a metaphor of parts. Despite this apparently obvious difference,
Socrates uses exactly the same language to describe them: he calls them ―types‖ or genē.
He uses an interesting phrase to describe the body in contrast with the soul: he says that
the shape of bodies ―take many genē‖: ―whether it [the soul] is one and homogeneous by
nature or takes many genē, like the shape of bodies…‖ (271a).80
The body has many
different genē, because it has many different shapes, which together compose the shape
of the body. These shapes are referred to as ―genē‖, which is strange given that one
would expect them to be called ―parts‖ rather than ―types‖.
However, when one considers ―types‖ and ―parts‖ relative to the Hippocratic
method being described, they are in many ways equivalent. Each genos is divided in
order to discover its dynamis or power relative to other things. On the one hand, different
―types‖ of personalities would be expected to react in similar ways to similar types of
speech, or different ―types‖ of liquid, for example, might be expected to react in similar
79
This is an important insight into a problem in scientific understanding per se: though we might be able to
use the scientific method to inform us of the causal interaction between particular events, one could not
even know what to look for without first appropriately dividing up one‘s subject so as to know what
possible causal interactions there might be. 80
πόηεξνλ ἓλ θαὶ ὅκνηνλ πέθπθελ ἠ θαηὰ ζώκαηνο κνξθὴλ πνιπεηδέο.
35
ways to different temperatures. On the other hand, different parts of a body would react
in different ways to different effects or, considered as organs, would themselves have
different powers or functions within a body. In both cases, whether one is considering
―types‖ or ―parts‖, one would need to have a proper division into types or parts so as to
begin a process of discovery of powers relative to different effects. In terms of their
relevance to the ―Hippocratic Method‖, then, what one might call ―types‖ or ―parts‖ are
largely equivalent, and it should not be too surprising that Socrates uses the same term,
―genē‖ to refer to both. In both cases, proper division of a subject into genē allows the
process of discovering the causal relationships necessary for the creation of a product.
In the Phaedrus, Socrates provides us with a detailed description of what he calls
the ―Method of Hippocrates‖ that underscores the importance of dialectic to any
productive craft. First, the collective part of dialectic provides crafts with an
understanding of their subjects as a whole, and without it, the craft would have no
product and be unable to seek any sort of end. Second, the divisive part of dialectic
allows the craftsperson to begin the process of discovering causal relationship between
types or parts, and without it, the craft would be unsystematic and incomplete. As a
result, any craft that intends to either be productive or understand causal relationships
requires dialectic in order to proceed.
7 Crafts and Measurement
Principle #5: Productive crafts are dependent on measurement to generate their
products.
The second meta-craft is measurement. Measurement has a special role in all
crafts, but particularly in productive ones. An extended example of the utility of
measurement for the military craft is provided in the Republic VII, though the
explanation for the connection is provided in the Philebus and the Statesman.81
Plato
81
Kenneth Dorter hypothesizes that the use of measurements in the crafts serves as a means of drawing the
philosopher from the realm of pure contemplation to the realm of practical action within the cave (Kenneth
Dorter, "Philosopher-Rulers: How Contemplation Becomes Action", Ancient Philosophy, 21 [2001]: 335-
56). Nonetheless, this is not the kind of utility I have in mind. I refer instead to the way in which the crafts
require measurement to function at all.
36
identifies two parts of the craft of measurement, which have an important role in any craft
in order for them to be ―most craft-like‖. The first art of measurement is purely
comparative measurement, through which one determines the relative value of particulars
on a continuum. The second craft of measurement is unit measurement, though which
one determines the values of an object relative to certain ratios or units. This second craft
of measurement takes slightly different shapes in the Philebus and in the Statesman: in
the Philebus, any type of measurement using units is considered a part of unit
measurement, but in the Statesman, unit measurement is always relative to an ideal
proportion. This unit measurement is especially important to productive crafts, as Plato
argues in the Philebus that the affixing of ratios using units is an important part of the
genesis or ―coming-to-be‖ of anything. In this section, I will detail the two types of the
art of measurement, and discuss the relevance of the arts of measurement to genesis,
especially with reference to medicine.
Socrates makes clear the need for mathematics in the crafts in Republic VII. In
that book, Socrates defends the teaching of mathematics to the military in his kallipolis.82
Socrates clearly claims that all crafts and knowledge require calculation:
That inconsequential matter of distinguishing the one, the two
and the three. In short, I mean number and calculation, for isn‘t
it true that every craft and body of knowledge must have a share
in that? (522c)83
Curiously, in order to defend this claim, Socrates does not provide a long list of
examples, as one might expect, but rather an extended example from war. He shows
several different ways in which soldiers use mathematics in their field.84
First, he cites an
example of Palamedes bragging about his contribution to war through the development of
numbers: ―He [Palamedes] says that, by inventing numbers, he established how many
troops there were in the Trojan army and counted their ships and everything else…‖
82
This passage in the Republic is at least in part comedic. However, Plato often uses comic passages to
elucidate important principles. 83
Σὸ θαῦινλ ηνῦην, ἤλ δ‘ ἐγώ, ηὸ ἕλ ηε θαὶ ηὰ δύν θαὶ ηὰ ηξία δηαγηγλώζθεηλ· ιέγσ δὲ αὐηὸ ἐλ θεθαιαίῳ
ἀξηζκόλ ηε θαὶ ινγηζκόλ. ἠ νὐρ νὕησ πεξὶ ηνύησλ ἔρεη, ὡο πᾶζα ηέρλε ηε θαὶ ἐπηζηήκε ἀλαγθάδεηαη αὐηλ
κέηνρνο γίγλεζζαη; 84
The use of war examples shows the Plato does not intend to limit the need for mathematics to productive
crafts. War is described as an acquisitive craft at Sophist 222c.
37
(522c-d).85
Geometry along with arithmetic is required in order to organize one‘s own
troops into formations and even to successfully set up a military camp, one requires
geometry:
Insofar as it pertains to war, it‘s obviously appropriate, for when
it comes to setting up camp, occupying a region, concentrating
troops, deploying them, or with regard to any of the other
formations an army adopts in battle or on the march, it makes all
the difference whether someone is a geometer or not. (526d)86
Finally, knowledge of the especially mathematical craft of astronomy is needed in order
to understand the seasons, months and years:
And what about astronomy? Shall we make it the third? Or do
you disagree?
That‘s fine with me, for a better awareness of the seasons,
months, and years is no less appropriate for a general than for a
farmer or navigator. (527c-d)87
Though Socrates and Glaucon aren‘t specific here as to the utility of knowing seasons,
months and years to a general, they are likely making reference to the obvious need to
know in which direction one is travelling as well as such less obvious needs such as
knowing the date so as to properly meet with other armies and be aware of possible
weather difficulties in a battle.
Though the connection between measurement and crafts is stated in the Republic,
the connection is explained in more detail in the Philebus and in the Statesman. Two
types of measurement are defined in the Philebus. The first sort is a purely comparative
kind of measurement, not involving any units at all. The second includes ratios between
different lengths:
SOCRATES: Whatever seems to us to become ‗more and less‘, or
susceptible to ‗strong and mild‘ or to ‗too much‘ and all of that
kind, all that we ought to subsume under the genus of the
unlimited as its unity…But look now at what does not admit of
these qualifications, but rather their opposites, first of all ‗the
85
ἠ νὐθ ἐλλελόεθαο ὅηη θεζὶλ ἀξηζκὸλ εὑξὼλ ηάο ηε ηάμεηο ηῶ ζηξαηνπέδῳ θαηαζηῆζαη ἐλ Ἰιίῳ θαὶ
ἐμαξηζκῆζαη λαῦο ηε θαὶ ηἆιια πάληα. 86
Ὅζνλ κέλ, ἔθε, πξὸο ηὰ πνιεκηθὰ αὐηνῦ ηείλεη, δῆινλ ὅηη πξνζήθεη· πξὸο γὰξ ηὰο ζηξαηνπεδεύζεηο θαὶ
θαηαιήςεηο ρσξίσλ θαὶ ζπλαγσγὰο θαὶ ἐθηάζεηο ζηξαηηᾶο θαὶ ὅζα δὴ ἄιια ζρεκαηίδνπζη ηὰ ζηξαηόπεδα ἐλ
αὐηαῖο ηε ηαῖο κάραηο θαὶ πνξείαηο δηαθέξνη ἂλ αὐηὸο αὑηνῦ γεσκεηξηθόο ηε θαὶ κὴ ὤλ. 87
Σί δέ; ηξίηνλ ζκελ ἀζηξνλνκίαλ; ἠ νὐ δνθεῖ;
κνὶ γνῦλ, ἔθε· ηὸ γὰξ πεξὶ ὥξαο εὐαηζζεηνηέξσο ἔρεηλ θαὶ κελλ θαὶ ἐληαπηλ νὐ κόλνλ γεσξγίᾳ νὐδὲ
λαπηηιίᾳ πξνζήθεη, ἀιιὰ θαὶ ζηξαηεγίᾳ νὐρ ἥηηνλ.
38
equal‘ and ‗equality‘ and, after the equal, things like ‗double‘,
and all that is related as number to number or measure to
measure: If we subsume all these together under the heading of
‗limit‘, we would seem to do a fair job. (24e-25a)88
Note that Plato is not talking about measuring something with a ruler, determining that
one stick is six and the other eight inches, and then concluding that the eight-inch stick is
larger than the six-inch stick because the number eight is greater than the number six.
This would involve ―comparing measure to measure‖, which will fall under the second
art of measurement. Rather, this part of the art of measurement seems like it may be
some sort of sensory power, as one is simply aware of the comparative sizes without
using anything like multiples. This sensory power would allow me to say, ―this car is
faster than that car‖ or ―this cat is larger than that cat‖. However, when I do this, I have
not yet reasoned about ―how much‖ faster or larger the one is than the other. To do this
requires more reasoning, specifically the mathematical reasoning that is involved in
ratios.
The second part of this art of measurement, as described in the Philebus
specifically includes the awareness of these ratios. Any measurement involving number
would involve the use of ratios. For instance, in order to determine a wall‘s length, I
might count the number of paces or feet that would be required to go from one end to the
other, thereby determining the multiple of one object (my foot) to the other (the wall).
Rulers or other measuring devices allow us to skip steps, as some of the measurement has
been done for us on the ruler itself. However, the ruler could not function without having
done this work of laying one ―inch‖ next to another so that we can measure the multiples
of one object by another. Other measurements are more complex, but require a similar
use of ratios. Measuring velocity requires measuring time and distance and, since the
development of special relativity, an inertial plane to treat as stationary. Before the
discovery of absolute zero, temperature was difficult to find units for, but discovery of
88
{Ω.} πόζ‘ ἂλ κῖλ θαίλεηαη κᾶιιόλ ηε θαὶ ἥηηνλ γηγλόκελα θαὶ ηὸ ζθόδξα θαὶ ξέκα δερόκελα θαὶ
ηὸ ιίαλ θαὶ ὅζα ηνηαῦηα πάληα, εἰο ηὸ ηνῦ ἀπείξνπ γέλνο ὡο εἰο ἓλ δεῖ πάληα ηαῦηα ηηζέλαη, θαηὰ ηὸλ
ἔκπξνζζελ ιόγνλ ὃλ ἔθακελ ὅζα δηέζπαζηαη θαὶ δηέζρηζηαη ζπλαγαγόληαο ρξῆλαη θαηὰ δύλακηλ κίαλ
ἐπηζεκαίλεζζαί ηηλα θύζηλ, εἰ κέκλεζαη.
{ΠΡΩ.} Μέκλεκαη.
{Ω.} Οὐθνῦλ ηὰ κὴ δερόκελα ηαῦηα, ηνύησλ δὲ ηὰ ἐλαληία πάληα δερόκελα, πξηνλ κὲλ ηὸ ἴζνλ θαὶ
ἰζόηεηα, κεηὰ δὲ ηὸ ἴζνλ ηὸ δηπιάζηνλ θαὶ πᾶλ ὅηηπεξ ἂλ πξὸο ἀξηζκὸλ ἀξηζκὸο ἠ κέηξνλ ᾖ πξὸο κέηξνλ,
ηαῦηα ζύκπαληα εἰο ηὸ πέξαο ἀπνινγηδόκελνη θαιο ἂλ δνθνῖκελ δξᾶλ ηνῦην.
39
the rate of liquid expansion ultimately allowed the creation of the thermometer.89
In all
of these cases, the same point applies: measurement using units requires measurement
using ratios; one must select some perhaps arbitrary unit and then measure everything
else in proportion to it. Once this has been done, one is able to discuss the ratios
independently of the objects, and consider their relationship, while the first art of
measurement only allows the ability to determine a few sensory objects and their
comparative values.
In the Statesman, Plato discusses an additional element to this second art of
measurement: he considers it to be intrinsically normative. The contrast between the two
sorts of measurement is that between a non-normative type and a type of measurement
compared to a normative unit, the metrion:90
It‘s clear that we would divide the art of measurement, cutting it
in two just the way we said, positing as one part of it all those
sorts of expertise that measure the number, lengths, depths,
breadths and speeds of things in relation to what is opposed to
them, and as the other, all those that measure in relation to what
is in due measure, what is fitting, the right moment, what is as it
ought to be – everything that removes itself from the extremes to
the middle. (284e)91
This is a curious claim for Plato to make, as it has the strange effect of building
normativity into the art of measurement, something that sounds quite strange to the
modern ear, accustomed to thinking of the ―facts‖ of measurement as being distinct from
the ―values‖ of normativity. Plato, however, defends the addition of normativity in two
ways: first, he allows that certain crafts require normativity as a part of their definition,
and would in fact be unintelligible without it. Second, he argues that measurement with
89
Note that Plato may not have believed a unit of temperature to be possible. None of his examples of unit
measurement involve temperature, while temperature examples are quite common when he discusses
purely comparative measurement. 90
I will leave the term ―metrion‖ untranslated in the body of the text, though it is translated as ―due
measure‖ below. Note that ―metrion‖ is not the term that is translated as ―mean‖ in Aristotle‘s ethics. That
is the term ―meson‖, which is translated as ―middle‖ below. Though in this passage, Plato lists a number of
normative units (―due measure...fitting...the right moment...what is as it ought to be‖), at 284b he had
introduced this argument referring to all of these under the single term, ―metrion‖: (―νὕησ θαὶ λῦλ ηὸ πιένλ
αὖ θαὶ ἔιαηηνλ κεηξεηὰ πξνζαλαγθαζηένλ γίγλεζζαη κὴ πξὸο ἄιιεια κόλνλ ἀιιὰ θαὶ πξὸο ηὴλ ηνῦ
κεηξίνπ γέλεζηλ;‖). 91
{ΞΔ.} Γῆινλ ὅηη δηαηξνῖκελ ἂλ ηὴλ κεηξεηηθήλ, θαζάπεξ ἐξξήζε, ηαύηῃ δίρα ηέκλνληεο, ἓλ κὲλ ηηζέληεο
αὐηῆο κόξηνλ ζπκπάζαο ηέρλαο ὁπόζαη ηὸλ ἀξηζκὸλ θαὶ κήθε θαὶ βάζε θαὶ πιάηε θαὶ ηαρπηῆηαο πξὸο
ηνὐλαληίνλ κεηξνῦζηλ, ηὸ δὲ ἕηεξνλ, ὁπόζαη πξὸο ηὸ κέηξηνλ θαὶ ηὸ πξέπνλ θαὶ ηὸλ θαηξὸλ θαὶ ηὸ δένλ θαὶ
πάλζ‘ ὁπόζα εἰο ηὸ κέζνλ ἀπῳθίζζε ηλ ἐζράησλ.
40
relationship to some sort of metrion is required for any productive craft to be truly
productive. I will address both of these points in turn, in order to show how Plato‘s
normative measurement can constitute a proper craft under his use of the term.
First, Plato argues in other places that some crafts are necessarily normative, and
would be completely unintelligible without normativity. In the Sophist, Plato divides up
division into two kinds, a non-normative kind and a normative kind: ―In fact in what
we‘ve called discriminations one kind separates what‘s worse from what‘s better and the
other separates like from like‖ (226d). The normative kind of separation he calls
cleansing, and he includes in this a number of different crafts. Cleansing, as a craft, is
simply unintelligible without normativity; it makes no sense at all to say that something is
being cleaned without making normative claims about what is bad and therefore ought to
be removed. He includes a number of crafts as cleansing, some of which might sound
surprising: gymnastics and medicine92
cleanse the inside of the body, while bathing
cleanses the outside of the body; winnowing separates wheat from chaff; retributive
punishment can cleanse society of its bad members, and so forth. All of these crafts,
however, are simply unintelligible without considering them not simply as separation of
something into kinds, but also, declaring some of those kinds bad and actively removing
them. Plato therefore has the theoretical space for a normative sort of craft. Just as
cleansing is normative when distinguishing between good and bad, metrion measurement
is normative when it discriminates between appropriate and inappropriate.
However, the primary reason why Plato argues for a normative kind of
measurement is that measurement in productive crafts is always normative. The reasons
for this are detailed in the Philebus, in which Plato gives an account of generation or
genesis that is highly dependent on number and therefore measurement. In that dialogue,
he separates all things into kinds: unlimited, limit, mixtures (of limit and unlimited), and
causes:93
92
That medicine appears here under cleansing is suprising, as medicine is considered a productive rather
than cleansing craft in nearly every other context. One interpretive possibility is the the Visitor is speaking
loosely and that medicine uses cleansing (among other means) to produce health, but is not a type of
cleansing, per se. 93
There is currently a great deal of discussion about the proper interpretation of the four greatest kinds
described in the Philebus. As there is insufficient space to defend a position in this debate, I have adopted
the interpretation of Dorothea Frede in her introduction to Plato, Philebus, ed. Dorothea Frede
(Indianapolis: Hackett, 1993).
41
SOCRATES: As the first I count the unlimited, limit as the second,
afterwards in third place comes the being which is mixed and
generated out of those two. And no mistake is made if the cause
of this mixture and generation is counted as number four? (27b-
c)94
To briefly summarise Socrates‘ claim here, all things are divided into four types:
unlimited things, limit, mixtures of the unlimited and limit and causes of these mixtures.
By unlimited things, Socrates is referring to various continua, such as hot and cold, dry
and wet and so forth. Limit refers to the various mathematical proportions in which those
continua might be placed. Mixtures are what occur when the two are brought together,
such as in the temperature of an actual body. Causes are whatever it is that puts the limits
into the continua. An example follows:
Unlimited: Heat and Cold
Limit: The proportion of heat and cold appropriate
to the body.
Mixture: The actual correct temperature of this patient
here.
Cause: Medicine
Plato claims that everything may be divided into these categories; I will not address this
claim here. Instead, I will discuss how this four-fold division is relevant to productive
crafts.
Plato puts the productive crafts into the fourth category of causes. Plato argues
that a craft is only craft-like in as much as it is capable of using measurement in its
process of generation:
SOCRATES: But as to building, I believe that it owes its superior
level of craftsmanship over other disciplines to its frequent use
of measures and instruments which give it high accuracy.
(Philebus 56b)95
Since generation is defined as the putting of measurement into matter, and since
productive crafts are productive in so far as they are able to generate a product,
94
{Ω.} Πξηνλ κὲλ ηνίλπλ ἄπεηξνλ ιέγσ, δεύηεξνλ δὲ πέξαο, ἔπεηη‘ ἐθ ηνύησλ ηξίηνλ κεηθηὴλ θαὶ
γεγελεκέλελ νὐζίαλ· ηὴλ δὲ ηῆο κείμεσο αἰηίαλ θαὶ γελέζεσο ηεηάξηελ ιέγσλ ἆξα κὴ πιεκκεινίελ ἄλ ηη; 95
{Ω.} Σεθηνληθὴλ δέ γε νἶκαη πιείζηνηο κέηξνηο ηε θαὶ ὀξγάλνηο ρξσκέλελ ηὰ πνιιὴλ ἀθξίβεηαλ αὐηῇ
πνξίδνληα ηερληθσηέξαλ ηλ πνιιλ ἐπηζηεκλ παξέρεηαη.
42
productive crafts are successful and ―craftlike‖ in as much as they are capable of
measurement. Without successful measurement, they rely on guesswork:
SOCRATES: If someone were to take away all counting,
measuring, and weighing from the arts and crafts, the rest might
be said to be worthless.
PROTARCHUS: Worthless, indeed!
SOCRATES: All we would have left would be conjecture and the
training of our senses through practice and knack. (Philebus
55e)96
When something is produced, ratios are put into some subject. In the case of something
like medicine, something that is out of harmony is put back into harmony, and that
harmony is constituted by the appropriate, measurable properties in a sick body. This
process of putting shape into something shapeless or returning shape to something
corrupted is what Plato calls genesis or ―production‖. For instance, in designing a house,
one uses some sort of blueprint to form a house into a certain shape.97
When healing a
body, one looks for a particular measurement that has deviated from its metrion, and
return the body to that metrion; for instance, if a patient had hypothermia or a fever, the
patient should be returned to a point of homeostasis of approximately 37°C. Without
measurement, however, there would be no reliable way to ensure that one is shaping what
one is producing, nor would there be any reliable way to ensure that one is returning
one‘s subject to its metrion. Measurement is therefore required for any productive craft,
as Plato defines production itself as the process of putting measurements into matter.
96
{Ω.} Οἷνλ παζλ πνπ ηερλλ ἄλ ηηο ἀξηζκεηηθὴλ ρσξίδῃ θαὶ κεηξεηηθὴλ θαὶ ζηαηηθήλ, ὡο ἔπνο εἰπεῖλ
θαῦινλ ηὸ θαηαιεηπόκελνλ ἑθάζηεο ἂλ γίγλνηην.
{ΠΡΩ.} Φαῦινλ κὲλ δή.
{Ω.} Σὸ γνῦλ κεηὰ ηαῦη‘ εἰθάδεηλ ιείπνηη‘ ἂλ θαὶ ηὰο αἰζζήζεηο θαηακειεηᾶλ ἐκπεηξίᾳ θαί ηηλη ηξηβῇ, 97
One difficulty with Frede‘s interpretation is that it is difficult to explain the building of a house as a
blending of opposites in a subject. This may represent a problem with Frede‘s interpretation, a tension in
Plato‘s own thought, or it may fall under one of the things that are ―left‖ to discuss as the dialogue breaks
off mid-discussion at 67b. Perhaps in construction, the continua of large and small, and strong and pliant
are bound in proportions of size (large and small) of compound materials (strong and pliant) so as to create
a sturdy and functional building.
43
8 The Hierarchy of Crafts
Principle #6: Crafts are arranged in a hierarchy, in which tool-supplying crafts are
subordinate to those that use their tools.
Plato also had a great deal to say, not only about the intrinsic properties of crafts,
but also about the relationship of one craft to another. Crafts are interrelated, and they
are interrelated in a very specific way. Some crafts produce or capture the tools of other
crafts. As a result, crafts are ordered in a hierarchy. The higher-order crafts are those
that use the results of the lower-order crafts.98
There are three reasons for this: first, the
higher-order craft, by using the result of the lower-order craft, makes that result useful
and therefore good; second, the higher-order craftsperson is the only person qualified to
evaluate the results of the lower-order crafts; finally, the higher-order craft is required in
order to provide a blueprint or a function for the lower-order craft. The higher-order craft
directs the lower-order craft. In this section, I will discuss the various relationships
between the higher- and lower-order crafts in Plato‘s corpus.
In the Gorgias, Socrates ridicules Callicles‘ claim that the more intelligent person
ought to have a greater share of goods, ―Perhaps the cobbler should walk around with the
largest and greatest number of shoes on‖ (490e).99
This farcical image of the cobbler
futilely trying to wear and tripping over dozens of over-sized shoes carries an important
message: no matter how excellent the shoe, its goodness is somehow relational. The
cobbler has no use for excess or over-sized shoes. Even a ―good shoe‖ can be a ―bad
thing‖ in the wrong context. In fact, should he try to wear them, they would be obstacles;
they would fail to function even as shoes. Unfortunately, Plato does not use a consistent
vocabulary for the two types of goodness implied here. Since this distinction runs
through almost everything that Plato says about the hierarchy of crafts and since the
vocabulary can be inconsistent, some introduction is required. The first is the sense of
goodness in which an excellent shoe is good, regardless of whether or not it is worn and
98
I use the term ―result‖ here because Plato does not limit the hierarchy only to productive crafts. A hunter
(an acquisitive craftsperson) captures something for use by a cook, for example. The term ―ergon‖ has this
double meaning of result and product: for example, a rabbit is the ergon of hunting while a house is the
ergon of building. I will translate it therefore as ―result‖. 99
ηὸλ ζθπηνηόκνλ ἴζσο κέγηζηα δεῖ ὑπνδήκαηα θαὶ πιεῖζηα ὑπνδεδεκέλνλ πεξηπαηεῖλ.
44
whether or not someone is tripping over it. This is what would today be called
―attributive‖ goodness, a kind of goodness whose content depends entirely on what kind
of thing ―good‖ is modifying and whose conditions are entirely determined by that kind
of thing. Plato‘s use of the term ―aretē‖ (excellence or virtue) consistently has this
attributive meaning, though he also sometimes also used the term ―agathon‖ (good) in an
attributive way. When discussing this first kind of goodness, I will use the term
―excellent‖ and ―excellence‖ in order to mark the way in which the term ―aretē‖ always
has this sense. At other times, Plato uses the term ―agathon‖ in a predicative way, not
simply to refer to something‘s quality as a member of a kind but something‘s goodness
simpliciter. On occasion, he even has characters insist that only this latter sense of
―goodness‖ is true goodness. This can create some confusion and even apparent
contradictions. Plato both uses ―agathon‖ in attributive and predicative ways and insists
that only predicative goodness is true goodness.100
In the Euthydemus, Socrates argues that attributively good tools are not
necessarily predicatively good. A carpenter, for example, does not derive any benefit
from his or her tools unless he or she uses them: ―For instance, if a carpenter were
provided with all his tools and plenty of wood but never did any carpentry, could he be
said to benefit from their possession?‖ (280c-d).101
Something that is attributively good
requires something else to be made predicatively good. Such things are evaluated for
their consequences. Further, excellent things are capable of actually being predicative
evils in the wrong hands: ―Now I suppose there is more harm done if someone uses a
thing wrongly than if he lets it alone - in the first instance there is evil, but in the second
neither evil nor good‖ (280e-281a).102
An excellent tool, then, can be predicatively good,
bad or neutral, depending on who uses it. As a result, in the Euthydemus, Socrates
ultimately denies that tools are predicatively good, no matter how excellent they might
100
I am using Peter Geach‘s ―attributive‖ goodness, in which the truth conditions of goodness are entirely
determined by the type of thing predicated from Peter Geach, "Good and Evil," in Theories of Ethics, ed.
Philippa Foot (New York: Oxford University Press, 1967), 64-73. Geach‘s attrubutive goodness allows me
to continue to reflect Plato‘s use of ―goodness‖ in both cases, while accurately capturing the distinction.
The alternative, which is to distinguish between ―excellence‖ and ―goodness‖ would provide the
impression that Plato does not sometimes use the terms synonyously, which he does. 101
νἷνλ ηέθησλ, εἰ παξεζθεπαζκέλνο εἴε ηά ηε ὄξγαλα ἅπαληα θαὶ μύια ἱθαλά, ηεθηαίλνηην δὲ κή, ἔζζ‘ ὅηη
ὠθεινῖη‘ ἂλ ἀπὸ ηῆο θηήζεσο; 102
πιένλ γάξ πνπ νἶκαη ζάηεξόλ ἐζηηλ, ἐάλ ηηο ρξῆηαη ὁηῳνῦλ κὴ ὀξζο πξάγκαηη ἠ ἐὰλ ἐᾷ· ηὸ κὲλ γὰξ
θαθόλ, ηὸ δὲ νὔηε θαθὸλ νὔηε ἀγαζόλ. ἠ νὐρ νὕησ θακέλ;
45
be. It is predicatively good if used by someone with wisdom, predicatively bad if used by
someone without wisdom, and neutral if used by no one:
So to sum up, Clinias, I said, it seems likely that with respect to
all the things we called good in the beginning, the correct
account is not that in themselves they are good by nature, but
rather as follows: if ignorance controls them, they are greater
evils than their opposites, to the extent that they are more
capable of complying with a bad master; but if good sense and
wisdom are in control, they are greater goods. In themselves,
however, neither sort is of any value. (281d-e)103
So, tools, which first appear to be predicative goods (in fact, Socrates and Clinias refer to
them as simply ―goods‖ until this point), are not really goods ―by nature‖, but only in
light of their consequences, which can be good or bad.
This list of goods in the Euthydemus that are tools and therefore are not
necessarily predicatively good includes health: ―...wealth and health and beauty - was it
knowledge that ruled and directed our conduct in relation to the right use of all such
things as these, or some other thing?‖ (281b).104
Health and the body of which it is an
attributively good state are tools for use by wisdom that are only made good by good use.
Plato presents this position most clearly in the (doubted) Alcibiades I, where he has
Socrates argue that the body is used by the body in the manner of a tool: ―Doesn‘t a man
use his whole body, too?‖ (129e).105
Even if the dialogue is not genuinely Platonic, the
approach is. Because the body can be either used or misused, depending on the wisdom
of the user, the body is not necessarily predicatively good. Rather, the body is only
predicatively good in so far as it is used properly. Even its attributively good state,
health, can be bad in the hands of a ―bad master.‖ The body is a tool that can be used or
misused, depending on a person‘s wisdom.106
The property of a tool that makes an attributively good tool only attributively
good is that it may be used for different ends. As a result, a particular tool can be called
103
λ θεθαιαίῳ δ‘, ἔθελ, ὦ Κιεηλία, θηλδπλεύεη ζύκπαληα ἃ ηὸ πξηνλ ἔθακελ ἀγαζὰ εἶλαη, νὐ πεξὶ
ηνύηνπ ὁ ιόγνο αὐηνῖο εἶλαη, ὅπσο αὐηά γε θαζ‘ αὑηὰ πέθπθελ ἀγαζὰ [εἶλαη], ἀιι‘ ὡο ἔνηθελ ὧδ‘ ἔρεη· ἐὰλ
κὲλ αὐηλ γῆηαη ἀκαζία, κείδσ θαθὰ εἶλαη ηλ ἐλαληίσλ, ὅζῳ δπλαηώηεξα ὑπεξεηεῖλ ηῶ γνπκέλῳ θαθῶ
ὄληη, ἐὰλ δὲ θξόλεζίο ηε θαὶ ζνθία, κείδσ ἀγαζά, αὐηὰ δὲ θαζ‘αὑηὰ νὐδέηεξα αὐηλ νὐδελὸο ἄμηα εἶλαη. 104
πινύηνπ ηε θαὶ ὑγηείαο θαὶ θάιινπο, ηὸ ὀξζο πᾶζη ηνῖο ηνηνύηνηο ρξῆζζαη ἐπηζηήκε ἤλ γνπκέλε θαὶ
θαηνξζνῦζα ηὴλ πξᾶμηλ, ἠ ἄιιν ηη; 105
{Ω.} Οὐθνῦλ θαὶ παληὶ ηῶ ζώκαηη ρξῆηαη ἅλζξσπνο; 106
I will return to some of the implications of this for autonomy in Chapter Four.
46
―good‖ in so far as it was used to do something good, successfully and skillfully and
―bad‖ in so far as it was used to do something evil, unsuccesssfully or ineptly. This
includes basic tools, like knives, and even more important tools like beauty, strength and
wealth, the fungible tool par excellence. When evaluating whether something was
predicatively good for someone, one must examine its consequences, rather than just its
powers. For example, beauty can be used to secure a successful marriage or to advance
the career of a tyrant. In fact, anything which can have both good and bad consequences
can be evaluated according to these consequences.
Here I‘d like to answer one common-sense objection to Socrates‘ position. Isn‘t a
knife a good knife just because of its ability to cut, not because of how it is used? For
instance, isn‘t a pruning knife just as good a pruning knife in the hands of a serial killer
as in the hands of a master vine-dresser? In the Republic I, Plato discusses the excellence
of a pruning knife. A pruning knife has the excellence of a pruning knife in so far as it
has the power of a pruning knife: ―I think you‘ll understand what I was asking earlier
when I asked whether the function of each thing is what it alone can do or what it does
better than anything else‖ (353a).107
A knife has the same power in the hands of a master
chef as in the hands of a serial killer. Should we not then say that the knife has the same
goodness in either case? The answer depends on whether one is referring to attributive or
predicative goodness. In either case, a functional pruning knife would be equally
excellent or attributively good. In the hands of a vine-dresser, the knife is of great
benefit; in the hands of a serial killer, it does great harm. As such, a tool should be
considered predicatively good or bad by virtue of its use, not simply its abilities.108
A
pruning knife in the hands of a vine-dresser is a good thing, but a pruning knife in the
hands of a serial killer is a very, very bad one.
A higher-order craft contributes more to a lower-order craft than merely providing
goodness to its results through use. Higher-order crafts are the only source of correct
107
Νῦλ δὴ νἶκαη ἄκεηλνλ ἂλ κάζνηο ὃ ἄξηη ξώησλ, ππλζαλόκελνο εἰ νὐ ηνῦην ἑθάζηνπ εἴε ἔξγνλ ὃ ἂλ ἠ
κόλνλ ηη ἠ θάιιηζηα ηλ ἄιισλ ἀπεξγάδεηαη. 108
One can see here a precursor to Aristotle‘s distinction between excellence and activity. A person is
happy in so far as he or she actualises his or her excellences, not simply by having them. So too a tool is
good in so far as it actualises its excellences, not simply by having them. Unused tools are just that,
useless. Further, a tool used to do harm is not only not useful, it is bad. Plato here is drawing the
distinction that Aristotle will later draw in more detail. Goodness is something that is not merely a matter
of capacity, but of actuality.
47
judgement of the quality of the result of a lower-order craft. Crafts that result in tools are
incomplete; the forms of their results are provided by higher-order crafts. What the lower
order craft provides is the knowledge of how to produce or acquire the result whose form
supplied by the higher-order craft. As a consequence, crafts are not as separate as it
might at first seem. Crafts that result in tools are dependent even for their existence and
purpose on higher-order crafts. Plato thus provides a hierarchy of evaluation between
crafts. The higher-order crafts are the best judge of the success of the lower-order crafts.
Plato presents his position clearly in both the the Republic and the Cratylus. In
the Republic, Socrates considers the problem that, since only a flautist will play a flute,
the flautist is a better judge of a good flute than the flute maker:
It‘s wholly necessary, therefore, that a user of each thing has
most experience of it and that he tell a maker which of his
products performs well or badly in actual use. A flute-player, for
example, tells a flute-maker about the flutes that respond well in
actual playing and prescribes what kind of flutes he is to make,
while the maker follows his instructions. (601d-e)109
This section of the Republic, however, is concerned mainly with imitation, and so there is
not much further discussion of the relationship between supplier and user there. More
discussion appears in the Cratylus, in which Socrates is interested in arguing that the
dialectician is the best judge of language. As they are those who use language, they are
the best judge of how well a language is constructed. In order to arrive at this conclusion,
he uses an analogy from other crafts, specifically carpentry, lyre-making and weaving.
Carpenters produce shuttles for weavers to use, and therefore, a weaver is the best judge
of the quality of the shuttle:
SOCRATES: Now who is likely to know whether the appropriate
form of shuttle is present in any given bit of wood? A carpenter
who makes it or a weaver who uses it?
HERMOGENES: In all likelihood, Socrates, it is the one who uses
it. (390b)110
109
Πνιιὴ ἄξα ἀλάγθε ηὸλ ρξώκελνλ ἑθάζηῳ ἐκπεηξόηαηόλ ηε εἶλαη θαὶ ἄγγεινλ γίγλεζζαη ηῶ πνηεηῇ νἷα
ἀγαζὰ ἠ θαθὰ πνηεῖ ἐλ ηῇ ρξείᾳ ᾧ ρξῆηαη· νἷνλ αὐιεηήο πνπ αὐινπνηῶ ἐμαγγέιιεη πεξὶ ηλ αὐιλ, νἳ ἂλ
ὑπεξεηζηλ ἐλ ηῶ αὐιεῖλ, θαὶ ἐπηηάμεη νἵνπο δεῖ πνηεῖλ, ὁ δ‘ ὑπεξεηήζεη. 110
{Ω.} Σίο νὖλ ὁ γλσζόκελνο εἰ ηὸ πξνζῆθνλ εἶδνο θεξθίδνο ἐλ ὁπνηῳνῦλ μύιῳ θεῖηαη; ὁ πνηήζαο, ὁ
ηέθησλ, ἠ ὁ ρξεζόκελνο [ὁ] ὑθάληεο;
{ΔΡΜ.} Δἰθὸο κὲλ κᾶιινλ, ὦ ώθξαηεο, ηὸλ ρξεζόκελνλ.
48
The issue can be seen in the phrase ―appropriate form of shuttle‖. Since a weaver is the
one who knows what to do with a shuttle, the weaver is the one who can tell if a shuttle
will be able to assist in producing clothing. This is because the weaver is the one who
knows about fabric and clothing, and this knowledge is required to know what form in
the wood will produce fabric and clothing: ―When a craftsman discovers the type of tool
that is naturally suited for a given type of product, he must embody it in the material out
of which he is making the tool‖ (389c).111
In other words, it would seem that if someone
is to know what tool to supply, one must know the result of the tool. However, this is not
understood by one‘s own craft, but by the next-higher-order craft. For instance, in order
to know what tool will best result in clothing, one must know what clothing is. However,
clothing is the result of weaving, not of carpentry. Therefore, a carpenter would also be
required to know the function of weaving. It would seem that all tool-supplying crafts
are pushed into the position where they must know the results of higher-order crafts.
Plato divides the knowledge in a tool-producing craft between the tool producers
and the tool users. The ―knowing why‖ is entirely part of the higher-order craft. The
―knowing what‖ must be known by both, but must be provided by the higher-order craft.
The blacksmith may know what to make simply because the schematics or the function
are provided to him or her by the saddle-maker. The saddle-maker may even hand the
blacksmith a schematic or a sample and say, ―Make me this widget.‖112
The knowing
how is entirely the province of the lower-order craft. What the lower-order craft provides
is an understanding of the causal relationships necessary to supply an object of the given
form once presented. If the higher-order craftsman knew how to make the tool, he or she
would have no use for the lower-order craftsman. The why then is known only by the
higher-order craftsman (he or she knows what stirrups are for), the what is known by both
111
ηὸ θύζεη ἑθάζηῳ πεθπθὸο ὄξγαλνλ ἐμεπξόληα δεῖ ἀπνδνῦλαη εἰο ἐθεῖλν ἐμ νὗ ἂλ πνηῇ [ηὸ ἔξγνλ], νὐρ
νἷνλ ἂλ αὐηὸο βνπιεζῇ, ἀιι‘ νἷνλ ἐπεθύθεη. 112
J.E. Tiles raises the interesting possibility that not every user understands enough to ―dictate
specifications‖ to a supplier in J. E. Tiles, "Technē and Moral Expertise", Philosophy: The Journal of the
Royal Institute of Philosophy, 1984: 56. It may be the case that the user shows up with a function in mind,
but the supplier draws up the actual specifications based on those desired functions. The final product is
then evaluated on the basis of the function in mind. Unfortunately, Plato does not go into such detail what
knowledge exactly is supplied by the user to the producer in either the Cratylus or Republic X . I discuss
the different senses in which morphology and function may be dictated by a tool-user in my paper, Daniel
Bader, "Using Plato‘s and Aristotle‘s Craft Analogies to Understand Biological Relationships", Examining
Telelogy Conference, 26 March 2010.
49
(though it is provided by the higher-order craft), and the how is known only by the lower-
order craftsman (only the blacksmith actually knows how to make stirrups).
How What Why
Supplying Craft knowledge right belief113
X
Using Craft X knowledge knowledge
By dividing up knowledge in this way, one can see in what sense a lower-order craftsman
is able to make something without knowing the higher-order craft: they receive the form
of their result from the higher-order craft, but do not, strictly speaking, know what it is
for.114
This position seems to threaten the possibility of therapeutic crafts. Therapeutic
crafts are those crafts that have the care of something as their product. Plato‘s two
favourite examples of therapeutic crafts are shepherding and medicine. The former takes
care of sheep and the latter takes care of the body. However, if the things taken care of
also serve as tools of a higher-order craft, and only a higher-order craft can identify a
good instance of something, how can there be a craft whose explicit product is ―the well-
being of x‖? Since someone with a craft still requires the ability to identify an instance of
its product, and the product of these crafts just is the good of their subjects, shouldn‘t
these crafts be able to evaluate? And wouldn‘t this imply that therapeutic crafts are able
to evaluate as well as identify their products? Moreover, if the subject of a therapeutic
craft is to be used by a higher-order craft, their products may actually be in conflict.
Take the following example: veal cattle. Qua cattle, cows are not better off in little
cages. Qua tools of food production, they are.115
If cattle are tools of higher-order crafts,
113
Socrates claims that the producer of the tool, rather than having knowledge (“epistēmē‖) of the result
has right belief (―pistin orthēn‖) (Republic 601e7). This is not the term usually translated belief, (―doxa‖).
Rather, it is a term for something one believes because someone of someone else‘s authority. 114
Though I do not discuss it in detail here, this direction would even apply to the acquisitive crafts. A
cook tells a fisherman what sorts of fish are fit for cooking and the fisherman brings back those fishing
while throwing back the rest. 115
J.E. Tiles raises the possibility that something analogous might be the case for health in J. E. Tiles,
"Technē and Moral Expertise", Philosophy: The Journal of the Royal Institute of Philosophy, 1984: 60. For
example, in a warlike society, it might be better to create a citizenry that is twice as strong but lives half as
long. I discuss applications of this difficulty in Chapter Four. When a tool has multiple excellences, and
trade-offs must be made, the using craft is required to adjudicate.
50
one produces the best cattle when one sticks them in little cages. However, a cowherd
should know that this does not constitute ―care for‖ the cattle in the therapeutic sense. A
therapeutic craft, then, seems able to evaluate whether or not something is good
independently of higher-order crafts; in fact, this evaluation is so independent, that it
might even arrive at a different conclusion than the higher-order craft.116
Plato would
seem to be mistaken.
A solution may be found again in Plato‘s distinction between attributive and
predicative goodness. In the Republic I, Socrates drew the distinction between
―shepherding‖ and ―money-making‖. The first was taking care of sheep, and the second
was butchering the sheep or selling them for butchery. Taking care of sheep, the
therapeutic craft, was the craft of producing aretē or excellence in the sheep. If one
assumes that excellent sheep are predicatively good, then the paradox of the above
paragraph arises. However, if excellent sheep are not necessarily predicatively good,
then whether or how many or to what extent excellent sheep should be made and current
sheep should be made excellent is subject to direction by another craft. Excellent sheep
are not necessarily predicatively good. Shepherding does not contain in it any guiding
principle as to how many sheep ought to be made excellent or whether the resources
devoted to making excellent sheep might be better used elsewhere. If producing
excellent sheep is necessarily predicatively good, how many such sheep should be
produced? A world overrun by rogue shepherds, devoted unflinchingly to the excellence
of their sheep, trampling and devouring crops, blocking traffic and getting their sheep
stuck in every drainpipe would not be a particularly pleasant world. Moreover, even a
single excellent sheep might not be of any use to a society of industrious vegans, who
would refuse to eat the sheep, wear its wool and would look askance at the strange
shepherd who is constantly fawning over his ―pet‖ rather than accomplishing something
useful. The aretē of an animal or other therapeutic subject may be related to its
116
See, for instance, Republic 345c-d, where Socrates separates the well-being of the sheep for the uses to
which the shepherd might put them: ―You think that, insofar as he‘s a shepherd, he fattens sheep, not
looking to what is best for the sheep but to a banquet...[On the contrary,] Shepherding is concerned only to
provide what is best for that which it is set over, and it is itself provided with all it needs to be at its best
when it doesn‘t fall short in any way of being the craft of shepherding‖ (―ἀιιὰ πηαίλεηλ νἴεη αὐηὸλ ηὰ
πξόβαηα, θαζ‘ ὅζνλ πνηκήλ ἐζηηλ, νὐ πξὸο ηὸ ηλ πξνβάησλ βέιηηζηνλ βιέπνληα ἀιι‘...πξὸο ηὴλ εὐσρίαλ,..
ηῇ δὲ πνηκεληθῇ νὐ δήπνπ ἄιινπ ηνπ κέιεη ἠ ἐθ‘ ᾧ ηέηαθηαη, ὅπσο ηνύηῳ ηὸ βέιηηζηνλ ἐθπνξηεῖ—ἐπεὶ ηά γε
αὑηῆο ὥζη‘ εἶλαη βειηίζηε ἱθαλο δήπνπ ἐθπεπόξηζηαη‖).
51
usefulness, just as a healthy horse who would be better able to pull a plough, but it may
be completely independent of its usefulness, such as healthy cow who might be a terrible
veal cow. Since the predicative goodness of a tool is determined by its usefulness, and
not by its excellence, even a therapeutic craft may be subject to its result‘s evaluation by
a higher-order craft.
In this section, I have set out the various ways in which Plato creates a hierarchy
of crafts. Crafts are ranked according to a hierarchy of use. First, using crafts are
responsible for the predicative goodness of their tools, rendering them good when they
use them properly. Second, they are necessary to direct the tool-producing crafts by
providing them with a goal and by evaluating their products.
9 Chapter Conclusion
In this chapter, I have laid out several Platonic positions concerning the nature of
crafts and their relationship to one another. As one can see, the Platonic conception of
crafts is very complex. In the course of this chapter, I have picked out seven principles
that will be relevant in the application of the Platonic conception of crafts to issues in
biomedical ethics. They are as follows:
Principle #1: There are two sorts of crafts: productive and acquisitive.
Principle #2: Crafts and knowledge are coextensive.
Principle #3: Crafts are teachable.
Principle #4: Crafts are dependant on dialectic for a determinate subject matter and
to systematically approach causal relationships.
Principle #4a: Crafts have a determinate subject matter.
Principle #4b: Crafts depend on the understanding of causal relationships.
Principle #5: Productive crafts are dependent on measurement to generate their
products.
Principle #6: Crafts are arranged in a hierarchy, in which tool-supplying crafts are
subordinate to those that use the tools.
52
In this chapter, I have explained these principles, and given the Platonic arguments for
them. In future chapters, I will apply these principles to problems in biomedical ethics,
demonstrating how they can be used to resolve or create important new insights to those
problems.
53
C h a p t e r 2 : C r a f t s a n d C o n t r a r i e s
1 Introduction
Having set out the Platonic principles of crafts, I will now show their application
to various issues in ethics, ancient and modern. The first issue I will address is the
definitional problem of what medicine is. Since before the Hippocratic Oath was written,
physicians have expressed concern about being asked to use the techniques of their crafts
to destroy rather than produce health. This concern is alive today in the debates
concerning euthanasia, elective abortion and execution through lethal injections. This is
not only a linguistic debate about how the term ―medicine‖ ought to be used, but a
conceptual one. The two parts of the concept of a ―craft of healing‖ collide in ways that
threaten to make the concept itself untenable or may force the concept to shift to ―craft of
healing and harming‖ in order to remain tenable. This problem, which I will call the
―bivalence of crafts‖, greatly concerns both Plato and Aristotle. If the same body of
knowledge allows one to both create and destroy, it threatens the concept of a craft with a
single product.
The difficulty lies in Principle #2: Crafts and knowledge are co-extensive.
There is something odd about saying one knows ―how‖ to do something. Knowledge is
generally knowledge of true propositions, and the combination of the term ―knowledge‖
with the indirect interrogative ―how‖ raises some strange puzzles117
. Specifically, if
knowing the same propositions enables us both to heal and to harm, is there any coherent
way to not say that knowing how to harm and how to heal are not exactly the same
knowledge? If medicine is knowledge, and knowledge is of a set of true propositionss,
then medicine and poisoning are in part the same thing, as they are knowledge of some of
the same set of true propositions. And, if they are the same thing, then a craft of
medicine whose product is simply healing and not harming does not exist.
One might respond that medicine is the use of that knowledge to heal rather than
to harm, but this raises a further difficulty. If medicine is the use of this knowledge to
117
I do not intend to make a firm distinction, nor do I believe does Plato, between ―knowing how‖ and
―knowing that‖ in the way that Gilbert Ryle does in Gilbert Ryle, "Knowing How and Knowing That",
Proceedings of the Aristotelian Society, 46 (1946): 1-16.
54
heal, then medicine is not identical with that knowledge and ceases to be knowledge.
Moreoever, if medicine is not identical with the knowledge that enables us to harm or
heal, then it is difficult to say what we should call that knowledge.
As a result, the concept of a craft whose product is health and not disease may be
an incoherent concept. It appears that either one must accept that the same craft has two
distinct products or abandon the claim that medicine is knowledge. Both Plato and
Aristotle were aware of this difficulty; in fact, they may have been the first to articulate it.
However, both Plato and Aristotle did believe that one can indeed construct a concept of
a craft of medicine that is knowledge and whose product is health and not disease. To
date, however, there has been little substantive discussion of to what extent they think
this might be the case and the reasons they think it might be true. Their arguments have
both linguistic and substantive parts. Plato develops a mode of strict speaking and then
attempts to show that, strictly speaking, crafts always benefit their subject. Aristotle
develops essential and accidental predication and then attempts to show that, essentially,
crafts are never destructive. In this chapter, I will first develop the problem of craft
bivalence as it was understood by Plato and Aristotle and then elucidate their solutions to
the problem.
2 Texts
The bivalence of crafts is discussed in many different texts, but some deal with
the difficulty at length. I introduce those below:
Republic, Book I: In this book, Thrasymachus and Socrates develop an understanding of
crafts, strictly speaking, and also discuss the nature of therapeutic crafts.
Hippias Minor: In this dialogue, Socrates addresses the concern that crafts seem to be
capable of contrary effects, and considers the ramifications of this, on the supposition that
virtues are crafts.
55
Statesman: In this dialogue, the Eleatic Visitor develops more clearly the mode of strict
speaking discussed in the Republic and also discusses the relationship of mathematics to
the crafts.
Philebus: In this dialogue, Socrates provides a definition of ―causes‖ that would rule out
the possibility of crafts being destructive.
Nicomachean Ethics, Book VI: In this book, Aristotle divides the intellectual virtues,
including crafts. He develops an understanding of the bivalence of crafts so as to show
that the virtues are not crafts.
Metaphysics, Book Θ: In this book, Aristotle includes crafts under powers, arguing that
they are capable of contraries, but only accidentally, not essentially.
3 Plato on the Problem of Crafts and Contraries
The difficulty that faced Plato and Aristotle is the result of a relatively
uncontroversial fact. It would appear that the ability to heal the human body also gives
people the ability to harm it. Study in the workings of the human body gives the
physician the knowledge of how to poison or otherwise kill a patient efficiently and even
secretly. Socrates states this bivalence of medicine most clearly in the Republic I: ―And
the one who is most able to guard against disease is also most able to produce it
unnoticed?‖ (333e).118
A bodyguard, for example, is often the best assassin, as Caligula
discovered, and one needs to be a good shot to ensure one misses, as was demonstrated
by the death of Joan Vollmer.119
In this section, I will examine Plato‘s arguments as to
why this sort of bivalence of the crafts exists and then explain in detail how this relates to
the problem of saying that one ―knows how‖ to do something. I will examine the
arguments for bivalence in the Republic and Hippias Minor, and then argue that the best
118
Ἆξ‘ νὖλ θαὶ λόζνλ ὅζηηο δεηλὸο θπιάμαζζαη, θαὶ ιαζεῖλ νὗηνο δεηλόηαηνο ἐκπνηήζαο; 119
Caligula, like several other Roman emperors, was killed by the Praetorian Guard, whose duty was to
protect the emperor (Josephus, The Antiquities of the Jews, Books 18-19, trans. Louis H. Feldman
[Cambridge: Harvard University Press, 1965], 212-30). Joan Vollmer was allegedly shot by her husband in
a drunken game of William Tell (Brenda Knight, Women of the Beat Generation: The Writers, Artists and
Muses at the Heart of a Generation [Boston: Conari Press, 1996], 52-53).
56
way to understanding how such bivalence might function can be found by examining the
Phaedrus.
In the Republic I, Socrates gives one of the clearest examples of bivalence, though
he does not there provide an argument as to why exactly such bivalence occurs.
Polemarchus makes the suggestion that justice could perhaps be useful in financial
partnerships concerned with guarding goods:
In what joint use of silver or gold, then, is a just person a more
useful partner than the others?
When it must be deposited for safekeeping, Socrates. (333c)120
However, Socrates responds to this by pointing out that the person best able to guard
goods is also the best at stealing the goods:
If a just person is clever at guarding money, then, he must also
be clever at stealing it.
According to our argument, at any rate.
A just person has turned out then, it seems, to be a kind of thief.
(334a)121
This was not the use of justice for which Polemarchus was hoping, and so he abandons
his claim.122
What I will examine, however, is what Socrates may have meant by
―clever‖ (―deinos‖) in this passage. I will provide and discuss three different senses in
which someone who is ―clever at guarding money‖ might be said to be clever at stealing
it: opportunity, particular knowledge and general knowledge. A person who has been
assigned to guard a particular object would have the best opportunity to steal that object.
However, this doesn‘t sound like cleverness, exactly. Even the world‘s least clever guard
would have little difficulty stealing an object placed under his or her unique protection.
The second possibility is particular knowledge. In such a case, the person guarding
something would know the details of the security surrounding the object and would be
120
Ὅηαλ νὖλ ηί δέῃ ἀξγπξίῳ ἠ ρξπζίῳ θνηλῇ ρξῆζζαη, ὁ δίθαηνο ρξεζηκώηεξνο ηλ ἄιισλ;
Ὅηαλ παξαθαηαζέζζαη θαὶ ζλ εἶλαη, ὦ ώθξαηεο. 121
Δἰ ἄξα ὁ δίθαηνο ἀξγύξηνλ δεηλὸο θπιάηηεηλ, θαὶ θιέπηεηλ δεηλόο.
Ὡο γνῦλ ὁ ιόγνο, ἔθε, ζεκαίλεη.
Κιέπηεο ἄξα ηηο ὁ δίθαηνο, ὡο ἔνηθελ, ἀλαπέθαληαη. 122
Note that the difficulty for Polemarchus is partly the result of Principle #1: There are two sorts of
crafts: productive and acquisitive. The apparently independent craft of guarding finds itself a by-product
of the acquisitive craft of stealing.
57
best able to steal the object if he or she were not guarding the object alone or even if he or
she were not on duty at the time. Knowledge such as when guards trade shifts, lock
codes, locations of keys and so forth could be used in a robbery. The guard would
thereby be capable of using his or her knowledge as a part of an ―inside job‖. It was in
this way, for example, that security guard Lou Werner was able to assist in the Lufthansa
Heist in 1978, despite not actually being on duty at the time.123
Unlike the guard with
opportunity, some cleverness is involved here, or at least some knowledge. The
particular knowledge of the ―insider‖ allows him or her to assist in and even organize
thefts. The third way in which Socrates might mean that a clever guard is a clever thief is
that someone who has the craft of guarding would also be the best thief. Understanding
how security systems work in general would allow the security expert to assist in and
plan a robbery based on what he or she knows about how security systems usually work.
Socrates seems to have at least opportunity and general knowledge in mind in the
different examples he provides in order to strengthen his case about bivalence. Between
333e and 334b, he cites three different examples that are supposed to demonstrate the
principle: a boxer who is best able to land a blow is best able to block one, the one who is
best able to guard against disease is best able to produce it unnoticed, and a guard is best
able to steal an enemy‘s plans. The example of the guard who is best able to steal an
enemy‘s plans is ambiguous, as it is unclear whether or not Socrates is talking about a
corrupt enemy guard currently guarding the enemy‘s plans (which would be an example
of opportunity) or one of one‘s own clever guards who is somehow sent in to the enemy
camp to steal the plans (which would be an example of general knowledge).124
The
example of a boxer, however, is a clear instance of general knowledge. No special
opportunity or inside knowledge is gained by being an excellent boxer that would allow
one to best block against blows. The case of medicine seems to involve both knowledge
and opportunity. The physician is best able to produce disease by his or her
understanding of medicine; however, the physician is also best able to produce it
unnoticed because no one else is checking up on his or her work. The Republic passage
123
Nicholas Pileggi, Wiseguy: Life in a Mafia Family (New York: Pocket Books, 1987), 227-35. 124
The later cited example of Autolycus, Odysseus‘s grandfather, may imply the latter interpretation.
Autolycus was ―the most accomplished perjurer and thief in the whole world‖ (Odyssey XIX:394).
However, Autolycus is nowhere described as a guardian, only as a thief, and the passage from the Republic
334a-b may only be citing Autolycus as an example of a thief, not as an example of a guardian-as-thief.
58
therefore makes use of the problem of bivalence, but leaves several problems
unanswered, such as in what exactly the cleverness referred to consists and why Socrates
believes it exists.
The dialogue Hippias Minor explores the same issue and in this case provides
more detail to the argument that crafts are bivalent. Socrates starts with a slightly
different problem than craft bivalence, truth-telling and lying. The character of Hippias
tries to contrast the two protagonists of the Iliad and the Odyssey, Achilles and Odysseus:
―In these lines he clearly shows the way of each man, that Achilles is truthful and simple
and Odysseus is wily and a liar...‖ (365b).125
However, Socrates disagrees with this
claim, arguing that, in fact, the truth-teller and the liar are the same person:126
Do you see, then, that the same person is both a liar and truthful
about these things, and the truthful person is no better than the
liar? For, indeed, he is the same person and the two are not
complete opposites, as you supposed just now. (367c-d)127
He argues this because someone can only be sure to lie if he or she knows the truth:
If someone were to ask you what three times seven hundred is,
could you lie the best, always consistently say falsehoods about
these things, if you wished to lie and never to tell the truth.
(366e-367a)128
For example, if I am asked what two plus two equals, I cannot be sure to speak a
falsehood unless I know the answer is four so as not to accidentally answer ―four‖ in my
attempt to lie. Since knowledge of the same fact is what makes me able both to tell the
truth and to lie, my knowledge of the fact that ―two plus two equals four‖ is what gives
me the power to tell the truth and lie about what two plus two equals. Knowledge of the
fact ―two plus two equals four‖ is therefore the state corresponding both to the power of
telling the truth about and to the power of lying about what two plus two equals.
Therefore the liar and the truth-teller are the same person.
125
ἐλ ηνύηνηο δεινῖ ηνῖο ἔπεζηλ ηὸλ ηξόπνλ ἑθαηέξνπ ηνῦ ἀλδξόο, ὡο ὁ κὲλ Ἀρηιιεὺο εἴε ἀιεζήο ηε θαὶ
ἁπινῦο, ὁ δὲ δπζζεὺο πνιύηξνπόο ηε θαὶ ςεπδήο. 126
Socrates doesn‘t mean that the same person both lies and tells the truth. Rather, the same person is a
―liar‖ and a ―truth-teller‖ in the sense that he or she has the power to do both. 127
{—Ω.} ξᾷο νὖλ ὅηη ὁ αὐηὸο ςεπδήο ηε θαὶ
ἀιεζὴο πεξὶ ηνύησλ, θαὶ νὐδὲλ ἀκείλσλ ὁ ἀιεζὴο ηνῦ ςεπδνῦο; ὁ αὐηὸο γὰξ δήπνπ ἐζηὶ θαὶ νὐθ
ἐλαληηώηαηα ἔρεη, ὥζπεξ ζὺ ᾤνπ ἄξηη. 128
εἴ ηίο ζε ἔξνηην ηὰ ηξὶο ἑπηαθόζηα πόζα ἐζηί, πόηεξνλ ζὺ ἂλ κάιηζηα ςεύδνην θαὶ ἀεὶ θαηὰ ηαὐηὰ ςεπδῆ
ιέγνηο πεξὶ ηνύησλ, βνπιόκελνο ςεύδεζζαη θαὶ κεδέπνηε ἀιεζῆ ἀπνθξίλεζζαη.
59
Socrates applies this principle not simply to lying and telling the truth, but to all
crafts and sciences. Knowing a craft allows us to perform the craft badly:
Well, then. As to the soul that plays the lyre and the flute better
and does everything else better in the crafts and the sciences –
doesn‘t it accomplish bad and shameful things and miss the mark
voluntarily, whereas the more worthless does this involuntarily.
(375b-c)129
As in the case of lying, someone can only fail to do something voluntarily by having the
craft and knowing how to hit the mark. So, for instance, a poor physician might
involuntarily kill his or her patient, but only a good physician can be sure to kill a patient
deliberately. A poor physician might even accidentally save a patient while trying to kill
him or her. Similarly, a poor flute player might accidentally play a beautiful note while
trying to play badly. Socrates generalizes to all crafts from his claim that knowing the
truth gives someone the ability to lie, saying that knowing a craft gives the person the
ability to perform a craft badly, apparently using Principle #2: Crafts and Knowledge
are co-extensive to ground his analogy.
The difficulty with this argument is that Socrates‘ claims don‘t seem to follow
from his examples. The narrow range of examples he uses to begin his case are
insufficient to demonstrate his conclusions. In fact, his examples concerning lying are
insufficient even to show that only someone who knows the truth can voluntarily lie.
Here are two examples to show the difficulty. If I didn‘t know how many planets there
are, I might accidentally say the true answer of eight (especially if I mistakenly thought
there were nine). Because the range of possible answers is so small, it is easy to
involuntarily hit the right answer. However, if someone were to ask me the president of
Mongolia, I am highly unlikely to accidentally answer ―Nambaryn Enkhbayar‖ in a bid to
successfully lie. While the possible answers are not infinite, they are so immensely large
that ignorance would be almost one-hundred percent as effective at allowing me to lie as
knowledge would be.130
It is true that knowledge would be the only way to be absolutely
certain I would not accidentally guess ―Nambaryn Enkhbayar‖, but Socrates has claimed
129
Σί δέ; θηζαξηζηηθσηέξα θαὶ αὐιεηηθσηέξα θαὶ ηἆιια πάληα ηὰ θαηὰ ηὰο ηέρλαο ηε θαὶ ηὰο ἐπηζηήκαο,
νὐρὶ ἀκείλσλ ἑθνῦζα ηὰ θαθὰ ἐξγάδεηαη θαὶ ηὰ αἰζρξὰ θαὶ ἐμακαξηάλεη, δὲ πνλεξνηέξα ἄθνπζα; 130
In English, one cannot ―lie‖ unless one knows the truth. However, the Greek verb ―ςεύδεζζαη‖ can also
mean ―to say that which is untrue, whether intentionally or not‖ (H.G. Liddell and R. Scott, Greek-English
Lexicon with a Revised Supplement [Oxford: Clarendon Press, 1996], 2021).
60
that the ignorant person would often accidentally tell the truth: ―Don‘t you think the
ignorant person would often (―pollakis‖) involuntarily tell the truth when he wished to
say falsehoods…?‖ (367a, my emphasis).131
―Often‖ is a far cry from ―almost never‖.
Moreover, when applied to the crafts, Socrates‘ argument seems even more
absurd. It makes very little sense to say that one might accidentally build an airplane or
even a barn, let alone that one would do it ―often‖. While there have been several
accidental inventions,132
most of those inventions have been either not especially
complex to produce (such as vulcanized rubber) or only discoveries of the principle of the
invention (such as photography). While it might be nomologically possible that someone
might accidentally build an airplane, it will almost certainly never happen. As a result, it
would appear that Socrates‘ attempt to generalize to all crafts from lying leads to
implausible conclusions.
However, Socrates makes another claim in the Hippias Minor that may clarify
why such bivalence exists. Socrates argues that only someone with a craft can be said to
have the power not to produce the product, because only the person with a craft can
choose not to produce the product: ―But each person who can do what he wishes when he
wishes is powerful‖ (366c).133
This is the same definition of powerful given in the
Gorgias, where he denies that rhetors have power because they do not do what they want:
I say, Polus, that both orators and tyrants have the least power in
their cities, as I was saying just now. For they do just about
nothing they want to, though they certainly do whatever they see
most fit to do. (466d-e) 134
If someone lacks a craft, however, that person cannot be said to have the power not to
produce the product of that craft, because that person‘s desires are irrelevant. They will
not produce the product, whether they want to or not. The only time they will produce
their product is by accident, which again would be completely independent of whether
131
ἠ ὁ κὲλ ἀκαζὴο πνιιάθηο ἂλ βνπιόκελνο ςεπδῆ ιέγεηλ ηἀιεζῆ ἂλ εἴπνη ἄθσλ. 132
Quinine, the battery, vaccination, anesthetics, photography, synthetic dye, celluloid, rayon, insulin,
antihistamines, the contraceptive pill, L.S.D., artificial sweetener, safety glass, nylon, Teflon, aspirin,
vulcanized rubber, cyclosporine, Velcro, corn flakes, and Post-It notes, to name a few (Royston M.
Roberts, Serendipity: Accidental Discoveries in Science [Toronto: John Wiley & Sons, 1989]). 133
{—Ω.} Γπλαηὸο δέ γ‘ ἐζηὶλ ἕθαζηνο ἄξα, ὃο ἂλ πνηῇ ηόηε ὃ ἂλ βνύιεηαη, ὅηαλ βνύιεηαη. 134
θεκὶ γάξ, ὦ Πιε, ἐγὼ θαὶ ηνὺο ῥήηνξαο θαὶ ηνὺο ηπξάλλνπο δύλαζζαη κὲλ ἐλ ηαῖο πόιεζηλ
ζκηθξόηαηνλ, ὥζπεξ λπλδὴ ἔιεγνλ· νὐδὲλ γὰξ πνηεῖλ ὧλ βνύινληαη ὡο ἔπνο εἰπεῖλ, πνηεῖλ κέληνη ὅηη ἂλ
αὐηνῖο δόμῃ βέιηηζηνλ εἶλαη.
61
they want it or not. Only a person with the craft can be said to choose not to produce the
product and therefore to have the power not to produce the product, because only the
person who has the ability to produce the product could have done otherwise had he or
she wished.135
Adding choice provides some of the explanation of how it is that someone with a
craft is capable of contraries. Someone with a craft can choose not to perform the actions
that will produce the product. However, this doesn‘t necessarily imply that the person
with the craft is able to produce the opposite of the product. For example, let us say that
a physician notices that performing a particular action, say providing medication y, will
allow him or her to produce health in a particular ill patient. The physician can therefore
choose not to provide medication y, and therefore health will not be produced. However,
this doesn‘t necessarily imply that the physician can produce disease in an otherwise
healthy patient. A healthy patient would require some overt action to be made unhealthy,
such as being poisoned. This second case is not the same as the first case. In the first
case, the physician could choose not to cure in a way that a person who did not know to
provide medication y could not be said to choose not to cure. In the second case,
however, the physician would need some sort of knowledge of a causal connection so as
to overtly poison the patient. This isn‘t a case of simply choosing not to apply medicine.
It seems, rather, to be a case of choosing to apply poison. As a result, demonstrating that
a physician can choose not to heal a patient does not demonstrate that he or she can
therefore choose to harm a patient.
In order to find a plausible account of how it is that a physician is capable of
contraries by virtue of knowing the same set of things, one must look instead to the
Phaedrus and Principle #4b: Crafts depend on the understanding of causal
relationships. As was discussed in Chapter One, the craftsperson knows two different
things. First, he or she knows the product associated with the craft. It is knowledge of
this whole that allows the craftsperson to produce something good. Second, he or she
knows a web of causal interactions that allow the craftsperson to produce that end. He or
135
Richard D. Parry argues that crafts are capable of contraries because their products can be either good or
bad. This is useful in explaining why someone might choose to destroy rather than to produce, but it
doesn‘t explain the mechanism that allows crafts to be so capable. To supplement this, I introduce the two
mechanisms: choice and knowledge of causal relationships (Richard D. Parry, Plato's Craft of Justice
[Albany: State University of New York Press, 1996], 87-89).
62
she knows how everything affects the subject of his or her craft and, if the subject of his
or her craft is complex, knows how everything affects each part or type of his or her
subject. Once both of these things are known, the product and the causal relationships,
the person is said to ―have the craft‖ of medicine. Knowing both the end and the causes
leading to that end provides the physician the power and choice not to cure. By knowing
what actions would heal the patient, the physician is able to decide not to perform the
action that would cure the patient. As such, even knowing simply how to cure patients
would provide the physician with knowledge of when to abstain from treatment so as not
to cure them.
However, knowledge of the product and all of the causal relations would also
introduce the physician to knowledge of how to damage the body as well as cure it. The
description of the Hippocratic Method in the Phaedrus would introduce the physician to
all sorts of effects that different potential treatments would have on a patient, both
positive and negative. That causal knowledge that a physician understands may be used
independently of understanding the good of the body. Once a physician has
understanding of that causal knowledge, the physician will learn a number ways of
manipulating the body that may be used at times to harm rather than help the health of the
patient. For example, a physician might learn how to lower a temperature and how to
raise a temperature and also the safe range of a temperature for a human body (36-38°C).
The physician might choose to use the techniques of medicine to raise a hypothermic
body or lower a febrile body to the safe temperature, but he or she might use those exact
same techniques to lower a hypothermic body‘s temperature even lower or raise a febrile
body‘s temperature even higher. By separating in-principle knowledge of the ends of
medicine from the techniques of medicine, Plato opened up the possibility that they might
be separated in practice. Fully aware of the safe limits of the body with the techniques of
medicine in hand, the physician would be in the best position not only to heal a sick
body, but kill a healthy one. A physician could thereby use the techniques of medicine
that move a body to a healthy state to move the body away from that healthy state.
Moreover, when one considers the scope of what Socrates claims to be a part of
the ―Hippocratic Method‖ in the Phaedrus, it becomes evident that one would discover
many interactions that might not be useable in order to heal the body at all. In the case of
63
the manipulation of raising or lowering temperature, it is the misapplication of an
otherwise medical technique. However, understanding all of the causal relationships
would lead to the discovery of a number of relationships that are bad for health per se,
without merely being a misapplication. The following passage was discussed in detail in
Chapter One:
If, on the other hand, it takes many forms, we must enumerate
them all and, as we did in the simple case, investigate how each
is naturally able to act upon what and how it has a natural
disposition to be acted upon by what. (270d)136
This Hippocratic Method would include the discovery not only of what helped the body,
but also what would outright harm the body. It would discover that ice brings down a
fever but also that hemlock kills a person. The systematic understanding of causation
presented in the Phaedrus would include the affective disposition of each part of the body
to every possible influence, including those that would destroy it. Therefore, as a result
of the systematic understanding of causal relationships between the body and other
things, the physician would discover not only techniques that could be misapplied, but
also a number of ways of destroying the body that no one else would know.
Therefore, knowing how to cure someone appears to be the same power as
knowing how to injure someone. The first account in the Hippias Minor, in which
Socrates argues from the need for knowledge so as not to accidentally tell the truth
―often‖, overestimates the extent to which someone in ignorance might accidentally fall
into success in complex cases. However, the second account in the Hippias Minor, in
which Socrates argues that only a craftsperson may choose to fail, augments the earlier
argument through a definition of choice and its relationship to power: one can only really
be said to have the power to do something if one chooses to do it. The account of the
methodology of the crafts in the Phaedrus provides some further material for
understanding how craft bivalence might function. Not only might a physician choose to
withhold care in the way that someone who is not a physician can never be said to choose
to withhold care, but a physician would discover in the course of learning the causal
relationships in medicine both ways to misapply otherwise medical techniques and also
136
ἐὰλ δὲ πιείσ εἴδε ἔρῃ, ηαῦηα ἀξηζκεζάκελνλ, ὅπεξ ἐθ‘ ἑλόο, ηνῦη‘ ἰδεῖλ ἐθ‘ ἑθάζηνπ, ηῶ ηί πνηεῖλ αὐηὸ
πέθπθελ ἠ ηῶ ηί παζεῖλ ὑπὸ ηνῦ;
64
causal relationships that could not even be plausibly considered beneficial in any respect,
such as the effect of hemlock on the body, as a physician understands the passive powers
of the body with respect to all effects, including harmful ones.
4 Plato’s Solution
Despite this bivalence of craft knowledge, Plato provides arguments to maintain
both the position that crafts are capable of contraries and the position that a craft is only,
strictly speaking, the power to produce one of those two contraries. His solution has two
steps: first, there is a kind of ―strict speaking‖ that may be used in order to divide up
something‘s powers into those that are essential and those that are accidental. Second,
the structure of crafts makes their use for benefiting subjects or generating artifacts
ontologically primary over their alternative accidental uses of harming or destroying.
4.1 The Linguistic Space
Plato carves out the linguistic space in which he can separate the craft of healing
from the craft of poisoning. He adopts the phrase ―strictly speaking‖ (―akribōs‖), and
argues that crafts, ―strictly speaking‖, always benefit their subjects. 137
The use of the
phrase in the Republic is somewhat puzzling, but those puzzles are resolved in the
Statesman, in which a class is defined according to a strict criterion and other apparent
instances that lie outside this criterion are not, strictly speaking, members of a class.
Crafts, which are sorts of powers, are defined not according to what knowledge gives a
power, but rather according to what power a body of knowledge gives. This means that a
craft can be said to strictly speaking only benefit rather than harm. In this section, I will
examine some of Socrates‘ earlier uses of the term ―strict speaking‖ and then discuss
them in light of the definition provided in the Statesman.
The term ―strict speaking‖ appears in Book I of the Republic as an important part
of the debate. It is used twice in quick succession by Thrasymachus and then by
137
What I am translating as ―strictly speaking‖ is the adverb ―akribōs‖, literally ―precisely‖. However, the
term ―precisely‖ can mean ―equivalent to‖. Rather, the phrase ―strictly speaking‖ captures better the
meaning of this term when used to describe definitions. In addition, sometimes the prefix ―akribo-― is
attached to a word meaning ―speaking‖ (such as Thrasymachus‘s ―akribologēi‖ at Republic 340d) with no
change in meaning.
65
Socrates. First, Thrasymachus wishes to argue that a ruler, strictly speaking, can never
err. He does this by making a distinction between the craftsperson and what one might
call the ―profession‖. The craftsperson is a craftsperson exactly in so far as he or she has
knowledge of a craft; the ―profession‖ on the other hand, consists of those who have
some sort of official status. Being a craftsperson is a matter of degree, as one is more or
less a doctor, for example, in as much as one knows medicine, while being a professional
is a boundary condition, as a person is either considered to be a doctor or not. ―Strictly
speaking‖, Thrasymachus claims, a craftsperson can never err, even though a professional
might. In so far as a craftsperson errs he or she lacks craft knowledge and therefore is not
a craftsperson:138
I think that we express ourselves in words that, taken literally, do
say that a doctor is in error, or an accountant, or a grammarian.
But each of these, insofar as he is what we call him, never errs,
so that, according to the precise account (and you are a stickler
for precise accounts), no craftsman ever errs. (340d-e)139
Thrasymachus is invoking Principle #2: Crafts and knowledge are co-extensive. A
craftsperson is only a craftsperson exactly in so far as he or she has the relevant
knowledge. Therefore, strictly speaking, argues Thrasymachus, a ruler can never err.
Rulership is a craft and therefore, in so far as a professional ruler errs, he or she is not
truly a ruler.140
Analogously, a doctor cannot, strictly speaking, be said to err, since a
physician only errs in so far as he or she lacks knowledge of how to heal, that is, in so far
as he or she is not really a physician.
Socrates counters that, in this same strict sense, a craftsperson is taking care only
of the subject of his or her craft: ―Is a doctor in the strict sense, whom you mentioned
before, a money-maker or someone who treats the sick? Tell me about the one who is
really a doctor?‖ (341c, emphasis added).141
Notice that Socrates believes that he is using
138
According to John Lyons, this isn‘t simply something that Thrasymachus believes, but is in fact an
analytic part of Plato‘s use of Greek. To be a craftsperson is to perform a craft well (John Lyons,
Structural Semantics: An Analysis of Part of the Vocabulary of Plato [Oxford: Basil Blackwell, 1963],
174). 139
ἀιι‘ νἶκαη ιέγνκελ ηῶ ῥήκαηη νὕησο, ὅηη ὁ ἰαηξὸο ἐμήκαξηελ θαὶ ὁ ινγηζηὴο ἐμήκαξηελ θαὶ ὁ
γξακκαηηζηήο· ηὸ δ‘ νἶκαη ἕθαζηνο ηνύησλ, θαζ‘ ὅζνλ ηνῦη‘ ἔζηηλ ὃ πξνζαγνξεύνκελ αὐηόλ, νὐδέπνηε
ἁκαξηάλεη· ὥζηε θαηὰ ηὸλ ἀθξηβῆ ιόγνλ, ἐπεηδὴ θαὶ ζὺ ἀθξηβνινγῇ, νὐδεὶο ηλ δεκηνπξγλ ἁκαξηάλεη. 140
―Professional ruler‖ sounds odd, but what is meant is the person recognized to be ruler. 141
ὁ ηῶ ἀθξηβεῖ ιόγῳ ἰαηξόο, ὃλ ἄξηη ἔιεγεο, πόηεξνλ ρξεκαηηζηήο ἐζηηλ ἠ ηλ θακλόλησλ ζεξαπεπηήο; θαὶ
ιέγε ηὸλ ηῶ ὄληη ἰαηξὸλ ὄληα.
66
the term ―strict speaking‖ in the same way as Thrasymachus was earlier. At first,
Thrasymachus doesn‘t understand the point that Socrates is trying to make, arguing that a
shepherd only takes care of sheep so as to eventually eat them and so the purpose of
shepherding is meat:
You think that shepherds and cowherds seek the good of their
sheep and cattle, and fatten them and take care of them, looking
to something other than their master‘s good and their own.
(343b)142
Using the ―strict sense‖ of shepherding described earlier, Socrates argues that
Thrasymachus is mistaken and that the shepherd, strictly speaking, takes care of the
sheep and is not trying to profit from them in the form of meat:
Nor would you call wage-earning medicine, even if someone
becomes healthy while earning wages?
Certainly not.
Nor would you call medicine wage-earning, even if someone
earns pay while healing. (346b)143
What Socrates is invoking here is Principle #4a: Crafts have a determinate subject
matter. While it may be true that a particular shepherd might eat his or her sheep, others
might sell the sheep, or shear the sheep or take care of the sheep as a pet. All of these,
however, count as shepherding. Therefore, the shepherd, strictly speaking, simply takes
care of sheep. The principle is perhaps even clearer in the case of the physician.
Someone might perform medicine for money or for honour or even for love. One might
even do it on a lark or under duress or from spite (to upset an unpleasant relative of the
patient). None of these further motivations, however, affects the definition of
―medicine‖. Strictly speaking, shepherding is taking care of sheep and medicine is taking
care of bodies.
However, the conception of ―strictly speaking‖ in these two passages remains
unclear, and it is not immediately obvious that the term is being used in the same way. In
the first instance, Thrasymachus argues that, strictly-speaking, a doctor never errs. In the
142
Ὅηη νἴεη ηνὺο πνηκέλαο ἠ ηνὺο βνπθόινπο ηὸ ηλ πξνβάησλ ἠ ηὸ ηλ βνλ ἀγαζὸλ ζθνπεῖλ θαὶ
παρύλεηλ αὐηνὺο θαὶ ζεξαπεύεηλ πξὸο ἄιιν ηη βιέπνληαο ἠ ηὸ ηλ δεζπνηλ ἀγαζὸλ θαὶ ηὸ αὑηλ. 143
Οὐδέ γ‘, νἶκαη, ηὴλ κηζζσηηθήλ, ἐὰλ ὑγηαίλῃ ηηο κηζζαξλλ.
Οὐ δῆηα.
Σί δέ; ηὴλ ἰαηξηθὴλ κηζζαξλεηηθήλ, ἐὰλ ἰώκελόο ηηο κηζζαξλῇ;
67
second instance, Socrates argues that, strictly-speaking, medicine benefits the bodies of
patients. Nonetheless, Socrates and Thrasymachus agree that ―strict-speaking‖ has the
same sense in both cases:
‗Consider this with the preciseness of language you mentioned.
Is it so or not?‘, I said.
‗It appears to be so‘, he replied. (342b)144
However, on the face of it, Thrasymachus and Socrates are using the term ―strictly
speaking‖ differently. Thrasymachus uses the term in order to signify whether or not
someone has successfully used his or her craft in order to produce its result or perhaps
whether or not one has the knowledge needed in order to produce its result. On the other
hand, Socrates uses the term on the basis of which result one ought to consider to be a use
of the craft. In fact, these seem to be quite separable. A professional physician might fail
to heal a patient, but still consider healing to be his or her goal as a physician.
Conversely, a physician might heal a patient successfully, but still mistakenly consider
his or her wages to have been the purpose of the action. Thrasymachus is introducing a
success condition, while Socrates is introducing a specificity condition145
. Why both of
them believe that the term ―strict speaking‖ has exactly the same sense in both cases is
unclear from the Republic.146
The same phrase appears again in the Statesman, this time in the context of
discovering what the true statesman is. As in the Republic, we do not receive a clear
definition of strictness of speech. We do, however, see this strictness of speech applied
in a systematic attempt to define the statesman. We are told the way in which one can be
said to be speaking strictly. What one looks for is a specific criterion by which one can
claim that something is a member of a particular class. Other things, no matter how
much they might resemble the class in question, cannot be considered a member of a
144
θαὶ ζθόπεη ἐθείλῳ ηῶ ἀθξηβεῖ ιόγῳ· νὕησο ἠ ἄιισο ἔρεη;
Οὕησο, ἔθε, θαίλεηαη. 145
Thrasymachus‘ success condition doesn‘t necessarily imply a specificity condition. Someone who has a
craft is successful at doing x. Socrates specifies this by limiting the scope of this success by limiting what
sort of x‘s would count as crafts if one is successful at them. 146
Kenneth Dorter argues that, because Thrasymachus introduced an idealised ruler who never errs,
Socrates was able to argue that an idealised ruler will also take care of sheep. However, this would appear
to beg the question, as Thrasymachus could have easily responded that his idealised ruler has an idealised
craft of exploitation (Kenneth Dorter, The Transformation of Plato's Republic [Oxford: Lexington Books,
2006], 38).
68
specific class unless they meet that criterion. It serves, then, to reduce the semantic field
of a term to a specific set. The way in which strictness of speaking is used in the
Statesman helps explain why Thrasymachus and Socrates may have been in agreement in
the Republic about the use of the term with reference both to success and to specificity.
In the Statesman, we are first introduced to the term ―strictness‖ (―akribeia‖)
early on in the dialogue, where the Visitor hopes that the confusion surrounding finding
the sophist from the prequel dialogue, the Sophist, might not be repeated when dealing
with the statesman. The Sophist had proven so difficult to find because he was an expert
imitator, and so appeared to be a great number of other things. The Statesman, on the
other hand, rather than appearing to serve so many other roles, rather has a great number
of imitators pretending at his or her throne. Strict speech may be used to clear out the
crowd of pretenders:
VISITOR: Our fears a little earlier were right, when we suspected
that we should prove in fact to be describing some kingly figure,
but not yet accurately to have finished the statesman off, until we
remove those who crowd round him, pretending to share his
herding function with him, and having separated him from them,
we reveal him on his own, uncontaminated with anyone else?
(268c)147
We aren‘t told yet how strict speaking will do this, but we have some more information
on what strict speaking can accomplish. In a case where many things resemble what one
is searching for, using strict speech can narrow down the semantic field.
In fact, the way in which the semantic field of the statesman is restricted in the
Statesman is substantially the same as the way in which the semantic field of justice is
restricted in the Republic. Two moves are made, one of which resembles the move made
by Thrasymachus, in which one should only consider something truly statesmanship if it
is made on the basis of knowledge, while another move is made resembling that of
Socrates, in which one should only consider rulership to be true rulership if it is for the
benefit of the ruled. In a way analogous to Thrasymachus‘ claim that a person is strictly
147
{ΞΔ.} Οὐθνῦλ ὀξζο ὀιίγνλ ἔκπξνζζελ ἐθνβήζεκελ ὑπνπηεύζαληεο κὴ ιέγνληεο κέλ ηη ηπγράλνηκελ
ζρῆκα βαζηιηθόλ, νὐ κὴλ ἀπεηξγαζκέλνη γε εἶκέλ πσ δη‘ ἀθξηβείαο ηὸλ πνιηηηθόλ, ἕσο ἂλ ηνὺο
πεξηθερπκέλνπο αὐηῶ θαὶ ηῆο ζπλλνκῆο αὐηῶ ἀληηπνηνπκέλνπο πεξηειόληεο θαὶ ρσξίζαληεο ἀπ‘ ἐθείλσλ
θαζαξὸλ κόλνλ αὐηὸλ ἀπνθήλσκελ;
69
only a ruler in so far as he has knowledge, the Eleatic Visitor argues that a constitution is
only a constitution if it is crafted with the knowledge of statesmanship:
VISITOR: It must be the case, it seems, that of constitutions too
the one that is correct in comparison with the rest, and alone a
constitution, is the one in which the rulers would be found truly
possessing expert knowledge… (293c, my emphasis)148
This mirrors Thrasymachus‘s claim in the Republic. There he argued that the ruler who
lacks knowledge of rulership isn‘t a ruler. Here a constitution written by a statesman
lacking knowledge of rulership isn‘t a constitution.
However, this is not the only criterion used to limit the semantic field of
―statesman‖. Just as Socrates added to Thrasymachus‘s claim that only a ruler who
knows what is to the advantage of the ruled is truly a ruler, so too does the Eleatic Visitor
add that the criterion for statesmanship is not simply that it is based on knowledge, but
that it is for the benefit of the ruled. He asks: ―[Is] the truest criterion of correct
government of a city the one according to which the wise and good man will govern the
interests of the ruled?‖ (296e) and receives an affirmative answer. 149
Just as in the
Republic, we first learn that correct rulership should only be considered rulership first if it
is from knowledge and second if it is for the benefit of the ruled. Also as in the Republic,
medicine is used as an analogy for this criterion. He asks rhetorically,
If then – to continue with our example – someone does not
persuade his patient, but has a correct grasp of the relevant craft,
and forces child, or man, or woman, to do what is better,
contrary to what has been written down, what will be the name
to give to this use of force? (296b-c, my emphasis)150
Notice the two phrases in italics above. The Stranger argues that the physician is still
acting as a physician so long as he or she is a) acting from craft knowledge and b) acting
for the health of the patient. These again are the criteria introduced by Thrasymachus and
Socrates respectively in the Republic. Thrasymachus argues that a doctor is only a doctor
148
{ΞΔ.} Ἀλαγθαῖνλ δὴ θαὶ πνιηηεηλ, ὡο ἔνηθε, ηαύηελ ὀξζὴλ δηαθεξόλησο εἶλαη θαὶ κόλελ πνιηηείαλ, ἐλ
ᾗ ηηο ἂλ εὑξίζθνη ηνὺο ἄξρνληαο ἀιεζο ἐπηζηήκνλαο. 149
ηνῦηνλ δεῖ θαὶ πεξὶ ηαῦηα ηὸλ ὅξνλ εἶλαη ηόλ γε ἀιεζηλώηαηνλ ὀξζῆο πόιεσο δηνηθήζεσο, ὃλ ὁ ζνθὸο
θαὶ ἀγαζὸο ἀλὴξ δηνηθήζεη ηὸ ηλ ἀξρνκέλσλ; 150
{ΞΔ.} Ἄλ ηηο ἄξα κὴ πείζσλ ηὸλ ἰαηξεπόκελνλ, ἔρσλ δὲ ὀξζο ηὴλ ηέρλελ, παξὰ ηὰ γεγξακκέλα ηὸ
βέιηηνλ ἀλαγθάδῃ δξᾶλ παῖδα ἢ ηηλα ἄλδξα ἠ θαὶ γπλαῖθα, ηί ηνὔλνκα ηῆο βίαο ἔζηαη ηαύηεο;
70
in so far as he or she knows how to heal a patient, while Socrates argued that a doctor is
acting as a doctor only in so far as he or she is healing the patient.
Strict speaking serves in both the Republic and the Statesman as a way of limiting
the semantic field so as to eliminate pretenders to a name. This limitation is done by
providing a specific criterion by which one determines what will and will not be properly
called by that name. A number of other things may imitate members of that group for a
number of reasons: pastry-bakers may imitate doctors or a tyrant may wear a crown.
However, once one sets the criteria according to which one divides up the world, these
apparent similarities become irrelevant. In the cases of both medicine and statesmanship,
characters in Platonic dialogues provide a criterion, that of knowledge that benefits a
subject. This use of ―strict speech‖ based on a criterion provides Thrasymachus, Socrates
and the Eleatic Stranger with the linguistic space in which to deny that something is a
member of a class. However, we do not yet know why this criterion is the proper
criterion for dividing off statesmen and doctors from their pretenders, nor do we know
why the Eleatic Stranger considers this a single criterion rather than two (knowledge and
benefit). To understand why this is, one must turn to the structure of crafts.
4.2 The Substantive Argument
In order to understand why knowledge for the benefit of a subject is not only a
criterion but the correct criterion for a craft, one must look at the structure of crafts and
causation themselves. Plato makes a number of claims about the nature of crafts and
causation that limit crafts only to those things that benefit their subjects. First, because
crafts have a power, they are not only knowledge, but also causes. Second, Plato limits
causes only to those things that provide ―genesis‖ or ―coming-to-be‖. Third, genesis
occurs only when one puts structure into something that might otherwise have been
unstructured. Finally, when one puts structure into something unstructured, one benefits
that subject. Therefore, crafts can only benefit their subjects, because they are causes that
put structure into their subjects rather than destroy them. As a result, the criterion of
knowledge that benefits a subject is not only an arbitrary criterion designed to make a
word more precise, but is actually a definition mandated by the nature of causation and
coming-to-be.
71
Part of the difficulty in understanding how there can be a craft of healing that is
not also identical with a craft of poisoning is that it appears that the same set of
information underpin both capacities: one uses the same set of information to heal a
patient as one might use to harm them. However, Socrates argues that powers are not
simply divided according to the conditions of a power, but according to what it is that the
power can do. In the Republic I, Socrates makes the claim that crafts ought to be divided
according to their powers:
Tell me, doesn‘t every craft differ from every other in having a
different power?...For example, medicine gives us health,
navigation gives us safety while sailing, and so on with the
others. (345e-346a)151
Notice what Socrates has done here. Crafts, despite being knowledge, are not divided up
according to which facts they contain, but what powers they give to someone. Take the
following example: knowing the fact that air will try to escape when pressure is increased
is both a part of the craft of fire-building (using a bellow) and also a part of proper
singing (exhalation). However, this does not imply that fire-building is partly singing nor
that singing is partly fire-building. Nor does it imply that fire-building and singing
―overlap‖. What it implies instead is that we can divide our knowledge according to the
powers given to us by having that knowledge and not according to the sets of facts that
underlie those powers. Fire-building and singing do not ―overlap‖ nor are they ―partly
the same‖. They are, instead, knowledge considered as powers and are divided up not
according to the particular facts but according to the powers.
How powers should be divided up is explained in a separate section of the
Republic that is not actually discussing crafts. In Book V, Socrates is concerned to
establish that knowledge and opinion are different because they are powers over different
subjects. In order to do this, he explains how it is that powers properly ought to be
divided:
A power has neither color nor shape nor any feature of the sort
that many other things have and that I use to distinguish those
things from one another. In the case of a power, I use only what
it is set over and what it does, and by reference to these I call
151
{ΞΔ.} Ἄλ ηηο ἄξα κὴ πείζσλ ηὸλ ἰαηξεπόκελνλ, ἔρσλ δὲ ὀξζο ηὴλ ηέρλελ, παξὰ ηὰ γεγξακκέλα ηὸ
βέιηηνλ ἀλαγθάδῃ δξᾶλ παῖδα ἢ ηηλα ἄλδξα ἠ θαὶ γπλαῖθα, ηί ηνὔλνκα ηῆο βίαο ἔζηαη ηαύηεο;…νἷνλ ἰαηξηθὴ
κὲλ ὑγίεηαλ, θπβεξλεηηθὴ δὲ ζσηεξίαλ ἐλ ηῶ πιεῖλ, θαὶ αἱ ἄιιαη νὕησ;
72
each the power it is: What is set over the same things and does
the same I call the same power; what is set over something
different and does something different I call a different one.
(477c) 152
Here Socrates is making interesting metaphysical claims about powers. Powers do not
have color or shape; they are invisible to the senses. Instead, powers can only be defined
by what they allow the person with the power a) to do and b) to what. So, for instance,
medicine can be defined as what allows someone a) to heal b) the body. In fact, since
crafts are powers, medicine and poisoning cannot be the same craft because they do not
do the same thing to the body. Although medicine and poisoning might include many of
the same facts, and although medicine and poisoning have the same subject in that they
are both ―set over‖ the body, they cannot be considered the came craft because they do
not share the same power. They simply do not do the same thing and therefore cannot be
considered the same craft despite the overlap in their factual bases.
However, Plato is not satisfied merely to divide up the crafts according to their
subjects and their powers. In fact, in many dialogues, characters assert that all crafts
benefit their subjects and that, in fact, there is only one craft per subject, that is, the one
that is set over it. For example, in the Euthyphro, we see a number of crafts divided up
according to what they care for: ―So dogs are benefited by dog breeding, cattle by cattle
raising, and so on with all the others. Or do you think that care aims to harm the object of
its care‖ (13b-c).153
However, in the Republic, makes a similar claim:154
Medicine doesn‘t seek its own advantage, then, but that of the
body?
Yes.
And horse-breeding doesn‘t seek its own advantage, but that of
horses. Indeed, no other craft seeks its own advantage – for it
152
δπλάκεσο γὰξ ἐγὼ νὔηε ηηλὰ ρξόαλ ὁξ νὔηε ζρῆκα νὔηε ηη ηλ ηνηνύησλ νἷνλ θαὶ ἄιισλ πνιιλ, πξὸο
ἃ ἀπνβιέπσλ ἔληα δηνξίδνκαη παξ‘ ἐκαπηῶ ηὰ κὲλ ἄιια εἶλαη, ηὰ δὲ ἄιια· δπλάκεσο δ‘ εἰο ἐθεῖλν κόλνλ
βιέπσ ἐθ‘ ᾧ ηε ἔζηη θαὶ ὃ ἀπεξγάδεηαη, θαὶ ηαύηῃ ἑθάζηελ αὐηλ δύλακηλ ἐθάιεζα, θαὶ ηὴλ κὲλ
ἐπὶ ηῶ αὐηῶ ηεηαγκέλελ θαὶ ηὸ αὐηὸ ἀπεξγαδνκέλελ ηὴλ αὐηὴλ θαι, ηὴλ δὲ ἐπὶ ἑηέξῳ θαὶ ἕηεξνλ
ἀπεξγαδνκέλελ ἄιιελ. 153
{Ω.} Καὶ νἱ θύλεο γέ πνπ ὑπὸ ηῆο θπλεγεηηθῆο, θαὶ νἱ βόεο ὑπὸ ηῆο βνειαηηθῆο, θαὶ ηἆιια πάληα
ὡζαύησο· 154
This also appears to be an application of Principle 4a: Crafts have a determinate subject matter.
Crafts can‘t share subject matters, which implies that somehow there is only one craft ―of‖ horses, which, it
turns out, is the one that benefits them.
73
has no further needs – but the advantage of that of which it is the
craft. (342c, my emphasis)155
In fact, without this exclusion, Socrates‘ argument in Book I simply would not work; it
would be possible that there be a craft of exploitation with the kind of success condition
that Thrasymachus had introduced, but that does not benefit the ruled.
Given the definition of powers later in Republic V, this might seem odd. Take
medicine and poisoning, for example. These would seem to be two separate powers
rooted in knowledge and therefore, two separate crafts. Each one has a different effect on
the same subject. They could be placed on the following table:
Craft Subject Product
Medicine Body Health
Poisoning Body Death
According to Socrates, powers are divided up by a) what they do b) to what. Why then is
a craft of poisoning, one that a) kills b) the body not possible? According to what was
said in Republic I though, such a craft is impossible and that was agreed by both Socrates
and Thrasymachus. If all crafts necessarily benefit their subject, poisoning not only isn‘t
a kind of medicine or a part of medicine, it isn‘t even a craft.
Unfortunately, no answer to this question is worked out in the Republic. Socrates
provides Thrasymachus with some examples, and Thrasymachus accepts them.
However, the solution to why all crafts must be beneficial is worked out instead in the
Philebus and in the Statesman. Craft knowledge in those dialogues has a specific
metaphysical status as a kind of cause, specifically, a kind of cause that fixes proportion
onto the material world. In that dialogue, Plato uses Principle #5: Crafts are dependent
on measurement to generate their products in order to explain why crafts are
necessarily beneficial. This is ―applied knowledge‖ in perhaps the most literal sense of
the term. One applies one‘s knowledge when one affixes the knowledge of correct
155
Οὐθ ἄξα, ἤλ δ‘ ἐγώ, ἰαηξηθὴ ἰαηξηθῇ ηὸ ζπκθέξνλ ζθνπεῖ ἀιιὰ ζώκαηη.
Ναί, ἔθε.
Οὐδὲ ἱππηθὴ ἱππηθῇ ἀιι‘ ἵππνηο· νὐδὲ ἄιιε ηέρλε νὐδεκία ἑαπηῇ—νὐδὲ γὰξ πξνζδεῖηαη—ἀιι‘ ἐθείλῳ νὗ
ηέρλε ἐζηίλ.
74
proportion in one‘s mind (one‘s understanding of form) and applies it to the material
world. However, when one affixes correct proportion into the material world, one
benefits the subject to which one has affixed the correct proportion. Therefore,
application of knowledge in the sense put forward in the Philebus and Statesman cannot
harm its subject. Instead, application of that knowledge in the sense of infusing
proportion into the world can only benefit. This means that poisoning is not truly a craft
as it destroys rather than creates proportion in the body.156
I return here to the discussion of the Philebus from Chapter One, focusing instead
on the role that measurement plays in generation or “genesis‖ within the crafts. In that
dialogue, Socrates divides all things into the limited, the unlimited, mixtures and the
causes of the mixtures. To recap briefly, the unlimited are the continua of various
properties in the material world. The limited are mathematical proportions of various
kinds. Mixtures are things in the world to which these mathematical proportions have
been affixed and cause is the craftsperson who applies those proportions to the world.
The imposition of these measurements is what allows more complex products to
come into existence. For instance, in the creation of a piece of music, the musician
weaves together high and low pitch in some sort of proportionate way, both temporally
through rhythm and simultaneously through harmony, to produce a piece of music.
When proportion arises, harmony arises, creating a piece of music:
And does not the same happen in the case of the high and the
low, the fast and the slow, which belong to the unlimited? Is it
not the presence of these factors in them which forges a limit and
thereby creates the different kinds of music in their perfection?‖
(Philebus 26a)157
A similar imposition of proportion can be seen in all the other crafts, Socrates argues,
which is what accounts for the importance of mathematics and precision in the crafts.
For example, housebuilders cut wood into geometrical shapes and align them in such a
156
One question this raises is whether or not there are any destructive crafts at all. It would seem that there
are not. Instead, apparently destructive crafts need to either be reframed in terms of something productive
or denied craft status. For instance, demolition of buildings would at first appear to be simply a destructive
craft, but is instead an instance of a cleansing craft, making a plot of land suitable for building or other use. 157
{Ω.} λ δὲ ὀμεῖ θαὶ βαξεῖ θαὶ ηαρεῖ θαὶ βξαδεῖ, ἀπείξνηο νὖζηλ, ἆξ‘ νὐ ηαὐηὰ [ἐγγηγλόκελα] ηαῦηα·
ἅκα πέξαο ηε ἀπεξγάζαην θαὶ κνπζηθὴλ ζύκπαζαλ ηειεώηαηα ζπλεζηήζαην;
75
way as to use their weight to form walls, balance and connect roofs to the top of houses.
If any of these proportions are incorrect or absent, the house is likely to collapse158
:
SOCRATES: But as to building, I believe that it owes its superior
level of craftsmanship to its frequent use of measures and
instruments, which give it high accuracy.
PROTARCHUs: In what way?
SOCRATES: In shipbuilding and housebuilding, but also in many
other woodworking crafts. For it employs straightedge and
compass, as well as a mason‘s rule, a line, and an ingenious
gadget called a carpenter‘s square. (56b-c)159
Crafts, according to Socrates, apply knowledge in a very direct way, by applying limit
directly into the world and imposing order so as to generate a product. If something did
not apply that order, one would not be generating anything and would not actually be
applying in this direct way that Socrates defines as craft-like.
Socrates explicitly limits the fourth category of things, causes, to crafts and every
example used in this section of the dialogue is a craft example. First, he identified
causing with production and producers with causes:
And is it not the case that there is no difference between the
nature of what makes and the cause, except in name, so that the
maker and the cause would rightly be called one? (Philebus
26e)160
.
Moreover, the craftsperson turns out to be another name for this fourth category of things,
causes: ―We therefore declare that the craftsman who produces all these must be the
fourth kind‖ (27b).161
A craftsperson, by definition, serves the metaphysical role as the
primary cause of the existence of things, and this role is satisfied only by the imposition
of proper proportion onto the unlimited continua.
158
We see here an application of Principle #5: Productive crafts are dependent on measurement to
generate their products. 159
{Ω.} Σεθηνληθὴλ δέ γε νἶκαη πιείζηνηο κέηξνηο ηε θαὶ ὀξγάλνηο ρξσκέλελ ηὰ πνιιὴλ ἀθξίβεηαλ αὐηῇ
πνξίδνληα ηερληθσηέξαλ ηλ πνιιλ ἐπηζηεκλ παξέρεηαη.
{ΠΡΩ.} Πῇ;
{Ω.} Καηά ηε λαππεγίαλ θαὶ θαη‘ νἰθνδνκίαλ θαὶ ἐλ πνιινῖο ἄιινηο ηῆο μπινπξγηθῆο. θαλόλη γὰξ νἶκαη
θαὶ ηόξλῳ ρξῆηαη θαὶ δηαβήηῃ θαὶ ζηάζκῃ θαί ηηλη πξνζαγσγίῳ θεθνκςεπκέλῳ. 160
{Ω.} Οὐθνῦλ ηνῦ πνηνῦληνο θύζηο νὐδὲλ πιὴλ ὀλόκαηη ηῆο αἰηίαο δηαθέξεη, ηὸ δὲ πνηνῦλ θαὶ ηὸ
αἴηηνλ ὀξζο ἂλ εἴε ιεγόκελνλ ἕλ; 161
{Ω.} Σὸ δὲ δὴ πάληα ηαῦηα δεκηνπξγνῦλ ιέγνκελ ηέηαξηνλ.
76
Imposing these limits can only be beneficial to the subject to which it is applied,
and Socrates uses the example of medicine to illustrate this point. For example, health is
the result when the opposites in the body are in the appropriate proportion: ―Is it not true
that in sickness the right combination of opposites establishes the state of health?‖
(Philebus 25e). 162
The claim is made even more clearly in the Laws III, in which he
directly attributes the lack of proportion to disease:163
If you neglect the rule of proportion and fit excessively large
sails to small ships, or give too much food to a small body, or too
high authority to a soul that doesn‘t measure up to it, the result is
disastrous. Body and soul become puffed up; disease breaks out
in the one, and in the other arrogance quickly leads to injustice.
(691c)164
As ―production‖ is identical with the imposition of proportion on something that would
otherwise be disproportionate, any craft that is set over something else will necessarily
benefit that subject. As a result, while powers are divided by a) what they are set over
and b) what they do to them, crafts ultimately are only divided by a) what they are set
over. Crafts only do one thing, which is to produce a beneficial order in their subjects.
Other powers cannot be considered crafts, and therefore, there is only one craft set over
each subject, the one that benefits it.
4.3 Conclusion
In order to establish that medicine as a craft does not include poisoning, Plato
makes two separate moves. First, he establishes the linguistic space for his conclusion,
referring to a method of ―strict speech‖ in which a term is used in such a way as to pick
out a unique criterion, in the case of medicine, the knowledge that gives the power to
benefit the body. Second, he establishes a substantive argument for the priority of the
healing craft over poisoning. As powers have no sensible properties, they are divided
162
{Ω.} Ἆξα νὐθ ἐλ κὲλ λόζνηο ηνύησλ ὀξζὴ θνηλσλία ηὴλ ὑγηείαο θύζηλ ἐγέλλεζελ; 163
Polybus, the likely author of the Hippocratic On the Nature of Man, makes a similar claim, indicating
that this view of health was not unique to Plato. He argues that, ―Health is primarily the state in which
these constituent substances are in correct proportion to each other, and are well mixed‖ (Hippocrates,
Hippocratic Writings, ed. G. E. R. Lloyd [Middlesex: Penguin Books Ltd., 1978], 262). 164
{ΑΘ.} άλ ηηο κείδνλα δηδῶ ηνῖο ἐιάηηνζη [δύλακηλ] παξεὶο ηὸ κέηξηνλ, πινίνηο ηε ἱζηία θαὶ ζώκαζηλ
ηξνθὴλ θαὶ ςπραῖο ἀξράο, ἀλαηξέπεηαί πνπ πάληα, θαὶ ἐμπβξίδνληα ηὰ κὲλ εἰο λόζνπο ζεῖ, ηὰ δ‘ εἰο ἔθγνλνλ
ὕβξεσο ἀδηθίαλ.
77
only by what objects they are set over and what they do to these objects. Second,
because crafts are causes and causes impose proportion on unlimited continua, and
because the imposition of correct proportion necessarily benefits the subject of that
proportion, all crafts are only crafts in so far as they benefit their subjects, and therefore,
poisoning and other types of destruction cannot be said to be crafts at all.
5 Aristotle on the Problem of Crafts and Contraries
Like Plato, Aristotle is aware of the problem of bivalence. Though Aristotle has a
different interpretation of bivalence than Plato, his approach to the solution is heavily
influenced by Plato‘s own. Examining his solution provides an example of the ways in
which Plato‘s discussion of the crafts, especially the medical craft, influences
contemporary debates. Further, the study of Aristotle‘s approach to a problem is always
helpful in understanding and clarifying Plato‘s own position. Aristotle notices as well
that those who have a craft seem best capable of producing the contrary of the product of
that craft, and cites medicine as his paradigmatic example. His clearest statement of the
principle occurs in Metaphysics Θ:
And each of those [powers] which are accompanied by reason is
alike capable of contrary effects, but one non-rational power
produces one effect; e.g. the hot is capable of heating, but the
medical art can produce both disease and health. (1046b5-7) 165
He and Plato approach the question differently, however. While Plato uses many
examples of bivalent crafts with few arguments, Aristotle uses much argumentation but
with few examples. In fact, one difficulty in interpreting Aristotle‘s passages is that one
must try to understand what kinds of examples he has in mind when providing his
argumentation. Throughout this section, I will consider two different senses in which
someone can be said to be ―capable of contraries‖:166
165
θαὶ αἱ κὲλ κεηὰ ιόγνπ πᾶζαη ηλ ἐλαληίσλ αἱ αὐηαί, αἱ δὲ ἄινγνη κία ἑλόο, νἷνλ ηὸ ζεξκὸλ ηνῦ
ζεξκαίλεηλ κόλνλ, δὲ ἰαηξηθὴ λόζνπ θαὶ ὑγηείαο. 166
In his commentary on Metaphysics Θ, Stephen Makin uses the phrases ―non-healing‖ and ―contra-
healing‖ to make a similar distinction (Aristotle, Metaphysics: Book [Theta], trans. Stephen Makin [New
York: Oxford University Press, 2006], 42).
78
1) The ―negative‖ interpretation: On this interpretation, the physician can be said to
cause disease by choosing not to heal a sick person and by choosing to abstain
from activities that would cause a healthy person from becoming diseased.
2) The ―positive‖ interpretation: On this interpretation, the physician can be said to
cause disease in the sense that he or she can generate disease in a patient who
would not have been diseased if not for the physician‘s interference.
Aristotle argues that a person with crafts is capable of contraries several times in the
Metaphysics, in Γ, Z and Θ. In this section, I will argue that, despite his debt to Plato
who clearly has a positive interpretation, the negative interpretation is the best
interpretation of Aristotle‘s claim that crafts are capable of contraries. In Γ, the negative
interpretation is the interpretation most compatible with the text. In Z and Θ, the text is
compatible with either interpretation, but a negative interpretation is the interpretation
that prevents the argument from being unsound.
Aristotle begins his discussion of crafts and contraries in Metaphysics Γ, his
―philosophical lexicon‖. Aristotle argues that crafts are capable of contraries because
they are principles and powers whose differentia is choice. Crafts appear in Γ under one
of his senses of ―principle‖ or ―archē‖, sometimes translated ―origin‖. The fifth sense of
an ―archē‖ lists crafts among them:167
That by whose choice that which is moved is moved and that
which changes changes, e.g. the magistracies in cities, and
dynasties and kingships and tyrannies, are called ‗principles‘,
and so are the crafts, and of these, especially the architectonic
arts. (1013a9-14)168
The term ―architectonic arts‖ here translates ―hai technai…architektonikai‖, a term that
includes both ―archē‖ and ―technē‖ in its very construction. Crafts appear here as a kind
of principle whose differentia is choice:
167
Christopher Kirwan claims that the intent of this fifth sense is to unite the middle-voiced ―archesthai‖,
meaning ―begin‖, with the active-voiced ―archein‖, meaning ―rule‖, though Aristotle doesn‘t himself take
any note of the different voices (Aristotle, Metaphysics, Books [Gamma], [Delta] and [Epsilon], ed.
Christopher Kirwan [Oxford: Clarendon Press, 1971], ad loc). 168
δὲ νὗ θαηὰ πξναίξεζηλ θηλεῖηαη ηὰ θηλνύκελα θαὶ κεηαβάιιεη ηὰ κεηαβάιινληα, ὥζπεξ αἵ
ηε θαηὰ πόιεηο ἀξραὶ θαὶ αἱ δπλαζηεῖαη θαὶ αἱ βαζηιεῖαη θαὶ ηπξαλλίδεο ἀξραὶ ιέγνληαη θαὶ αἱ ηέρλαη, θαὶ
ηνύησλ αἱ ἀξρηηεθηνληθαὶ κάιηζηα.
79
Because choice is involved in the use of a craft, the craft can be considered a principle
and, not only this, but a principle akin to other sorts of leadership. In this category,
Aristotle puts all the crafts. Though he says it is ―especially‖ (―malista‖) architechtonic
arts, by which he likely means people who command other craftspeople, all craftspeople
are included under this sense of principle. All crafts, therefore, are a kind of principle
because they allow actions that follow from choice.169
Crafts also appear under the heading of ―power‖ or ―dynamis‖. The first two
senses are connected, so I will cite them both. Crafts appear explicitly under the first
sense of power, which he defines as:
…a source of movement or change, which is in another thing or
in the same thing qua other, e.g. the art of building is a capacity
which is not in the thing built, while the art of healing, which is a
capacity might be in the man healed, but not in him qua healed.
(1019a15-18)170
Crafts are a kind of power as well as a kind of principle, in this case as a principle of
motion in another:171
169
Note also Plato‘s use of choice as a source of power in the Hippias Minor, as discussed above. 170
Γύλακηο ιέγεηαη κὲλ ἀξρὴ θηλήζεσο ἠ κεηαβνιῆο ἐλ ἑηέξῳ ἠ ᾗ ἕηεξνλ, νἷνλ νἰθνδνκηθὴ δύλακίο
ἐζηηλ ἡ νὐρ ὑπάξρεη ἐλ ηῶ νἰθνδνκνπκέλῳ, ἀιι‘ ἰαηξηθὴ δύλακηο νὖζα ὑπάξρνη ἂλ ἐλ ηῶ ἰαηξεπνκέλῳ,
ἀιι‘ νὐρ ᾗ ἰαηξεπόκελνο. 171
Kirwan notices an ambiguity in the “qua other‖ here: ―Is he doctored qua other (1) because his patient
does not have to be himself or (2) because his patient does not have to be a doctor?‖ (Aristotle,
Metaphysics, Books [Gamma], [Delta] and [Epsilon], ed. Christopher Kirwan [Oxford: Clarendon Press,
1971], ad loc). This is the result of a larger puzzle in this classification. Some changes can only take place
in oneself (growth), some can take place only in another (fire cannot heat itself) and some can take place in
either (healing). Why he chooses to combine the last two into a single category is a puzzle in its own right.
Principles
Move through Choice
GovernmentCrafts (especially
architectonic)
Other senses of “principle”
80
A power in this sense is a source of motion or change in some patient (or in the
craftsperson qua other, by which Aristotle is referring to cases where a doctor uses his or
her knowledge for self-healing). Notice that this sense of power would include non-
rational powers as well, despite all of Aristotle‘s examples being craft examples. Fire,
for example, is a source of movement or change in another thing. Crafts, then, are
powers of the sort that produce change in another, that is, they are a making or ―poiēsis‖.
However, since it would apply to any power, including non-rational powers that are not
capable of contraries, it cannot be qua powers in this first sense that crafts are capable of
contraries.
The second sense of ―power‖ ties together Aristotle‘s claim that crafts are
principles because they involve choice and his claim that crafts are powers because they
involve making. In the second sense, one can be said to have a ―power‖ because one has
―choice‖:
The capacity of performing this well or according to choice; for
sometimes we say of those who merely can walk or speak but
not well or not as they choose, that they cannot speak or walk.
(1019a23-25)172
Several things about this passage are worth noticing. This has similarities to the fifth
sense of ―principle‖ mentioned earlier. In the fifth sense of principle, a principle is ―that
by whose choice that which is moved is moved and that which changes changes‖. In the
second sense of power, it is ―the capacity of performing this [i.e. being a source of
movement or change, which is in another thing] well or according to choice‖. There are
some differences. The definition of ―principle‖ mentions ―others‖, while the definition of
172
ἔηη ηνῦ θαιο ηνῦη‘ ἐπηηειεῖλ ἠ θαηὰ πξναίξεζηλ· ἐλίνηε γὰξ ηνὺο κόλνλ ἂλ πνξεπζέληαο ἠ εἰπόληαο,
κὴ θαιο δὲ ἠ κὴ ὡο πξνείινλην, νὔ θακελ δύλαζζαη ιέγεηλ ἠ βαδίδεηλ· ὁκνίσο δὲ θαὶ ἐπὶ ηνῦ πάζρεηλ.
Powers
Source of change in another
CraftsOther sources of
change in another, such as heat
Changes in oneself, such as growth or
locomotion
81
―power‖ mentions ―well‖. Nonetheless, it appears that Aristotle means quite similar
things by his fifth sense of principle and his second sense of power: having the choice to
cause change. Second, notice that the two examples here, walking and speaking, are not
crafts. While in one sense, one needs to learn to walk and speak, one does not learn them
in the same way as one learns to practice medicine or build a house. Walking does not
even require reason; a newborn foal can walk. Though craft examples have recurred
throughout Aristotle‘s definitions of ―principle‖ and ―power‖, none of his definitions
have so far restricted themselves to crafts. Crafts are principles and powers not qua crafts
but qua including choice.
Choice, then, is a plausible place to look for an explanation of Aristotle‘s claim
that crafts are capable of contraries. After all, choice, among other things, is a choice
between contraries. One sense each of both principle and power is that which has choice
as its differentia. Nonetheless, while this might provide some explanation of why
Aristotle thinks that crafts are capable of contraries, it does not solve the problem of the
sorts of contraries of which a craftsperson is capable. After all, both the positive and
negative interpretations of bivalence include choices. However, there is more evidence in
Γ that implies the negative interpretation, rather than the positive interpretation. The first
piece of evidence is in his discussion of ―priority‖ and ―posteriority‖, which he defines
together. The fourth sense of ―priority‖ and ―posteriority‖ is priority and posteriority in
power, in which ―choice‖, ―power‖, ―motion‖ and ―principle‖ all appear173
:
Others are prior in power; for that which exceeds in power, i.e.
the more powerful, is prior; and such is that according to whose
choice (―prohairesin‖) the other – i.e. the posterior – must
follow, so that if the prior does not set it in motion the other does
not move, and if it sets it in motion it does move; and here
choice is a principle. (1018b21-26)174
This sense of priority has a great deal of overlap with the fifth sense of principle and the
second sense of power. One thing is said to be ―prior‖ to another in ―power‖ when the
―choices‖ of the first thing are a ―principle‖ of ―motion‖ in the second thing. Especially
173
Notice, too, just how well this definition maps onto both rulership and crafts, and Aristotle could have
just as easily provided a rulership example as a craft example. Curiously, it is the only sense of priority for
which he does not provide an example. 174
ηὰ δὲ θαηὰ δύλακηλ (ηὸ γὰξ ὑπεξέρνλ ηῇ δπλάκεη πξόηεξνλ, θαὶ ηὸ δπλαηώηεξνλ· ηνηνῦηνλ δ‘ ἐζηὶλ νὗ
θαηὰ ηὴλ πξναίξεζηλ ἀλάγθε ἀθνινπζεῖλ ζάηεξνλ θαὶ ηὸ ὕζηεξνλ, ὥζηε κὴ θηλνῦληόο ηε ἐθείλνπ κὴ
θηλεῖζζαη θαὶ θηλνῦληνο θηλεῖζζαη· δὲ πξναίξεζηο ἀξρή).
82
interesting for the purpose of understanding whether Aristotle intends a negative or
positive understanding of bivalence, is his characterization of the type of choice here.
Notice that Aristotle does not characterize the choice as being between different motions
that he or she might induce in a subject. Instead, the choice is characterized as being
between whether or not to set something in motion. In the case of the bivalence of crafts,
this favours the negative interpretation. If the positive interpretation were correct, then
there would be two motions that a physician might initiate, and three possible actions:
healing, harming and letting be. However, Aristotle characterizes choice as between
initiating a motion and not initiating a motion.
A further piece of evidence from Γ that Aristotle intends a negative interpretation
is in his definition of ―cause‖. Perhaps the most obvious objection to the negative
interpretation is that, when Aristotle claims that a physician can ―cause disease‖, it is
simply implausible to interpret this claim as ―does nothing and allows disease to happen‖,
so therefore the positive interpretation must be correct. However, Aristotle includes
almost exactly this in his definition of ―cause‖. Specifically, he describes a way in which
something can be ―of contraries‖ or ―tōn enantiōn‖, the exact same phrase used in Θ:175
Again, the same thing is sometimes the cause of contraries; for
that when present causes a particular thing, we sometimes
charge, when absent, with the contrary, e.g. we impute the
shipwreck to the absence of the steersman, whose presence was
the cause of safety; and both – the presence and the privation –
are causes that move. (1013b12-16) 176
This claim is perhaps the best evidence for a negative interpretation of bivalence in
Aristotle, and certainly the best evidence against the view that a positive interpretation is
the obvious interpretation of Metaphysics Γ. Here, Aristotle argues explicitly that the
same cause can be considered a cause of contraries, even a cause of contrary motions,
175
Kirwan argues that this paragraph also implies that we ought to blame a present steersman for a crash
qua (failed) steersman, because of the term ―aitiōmetha‖ may have the sense of ―blame‖, implying that
there is actually a steersman to blame for the accident (Aristotle, Metaphysics, Books [Gamma], [Delta]
and [Epsilon], ed. Christopher Kirwan [Oxford: Clarendon Press, 1971], ad loc). However, it is clear from
the passage that the steersman‘s absence is what causes the shipwreck, not his or her poor performance, and
if there is any blame, it would be for a steersman‘s failing to show up rather than failing to steer correctly.
On the contrary, the term ―aitiōmetha‖ may simply have the sense of ―we attribute causation‖, thereby not
implying that there is any particular steersman who is ―blameworthy‖ for the crash. 176
ἔηη δὲ ηαὐηὸ ηλ ἐλαληίσλ ἐζηίλ· ὃ γὰξ παξὸλ αἴηηνλ ηνπδί, ηνῦη‘ ἀπὸλ αἰηηώκεζα ἐλίνηε ηνῦ ἐλαληίνπ,
νἷνλ ηὴλ ἀπνπζίαλ ηνῦ θπβεξλήηνπ ηῆο ἀλαηξνπῆο, νὗ ἤλ παξνπζία αἰηία ηῆο ζσηεξίαο· ἄκθσ δέ, θαὶ
παξνπζία θαὶ ζηέξεζηο, αἴηηα ὡο θηλνῦληα.
83
both when it is present and when it is absent.177
Aristotle‘s own example is the lack of a
steersman causing a boat to crash. In fact, the claim is so strong here, it would apply to
non-rational as well as rational causes. For example, the absence of rain causes trees to
die or even the absence of sufficient gravity causes moons to escape orbit. So, one says
that the absence of rain causes trees to die because the presence of rain causes them to
flourish, and the absence of sufficient gravity causes the moon to escape orbit because the
presence of sufficient gravity causes moons to stay in orbit. Absences are causes of
motion exactly insofar as the presence of their positive term would have been a cause of
the opposite motion.
When one connects this characterization of absences as causes, a negative
interpretation of bivalence becomes plausible. A physician has the power to heal (the
first sense of ―power‖). However, because the physician is rational, he or she has the
choice whether or not to exercise that power. If the physician chooses to heal the patient,
then the doctor initiates the motion of healing in the patient. However, if the physician
chooses instead not to heal the patient, the patient remains ill. The absence of the activity
of healing from the physician when he or she decides not to heal the patient is analogous
to the absence of the steersman. As it is true that a ship can crash because it lacks a
steersman, it is also true that a ship can crash because the steersman chooses not to steer.
By choosing not to heal, a physician chooses an absence of healing, and that absence of
healing can be said to be the source of motion for the absence of whatever it is that
activity of healing would have caused had it been initiated. Thus, a person might say that
he or she is sick because no physician is willing to perform a cure, which is not
uncommon for highly expensive procedures. People with choices then are capable of
contraries in the sense that the absence of an activity can be considered the cause of the
absence of its effect.
In Metaphysics Z.7, Aristotle again discusses crafts and contraries, and again his
claim is best understood with a negative interpretation. Aristotle makes the point in Z
that understanding something and understanding its privation are exactly the same:
For even contraries have in a sense the same form; for the
substance of a privation is the opposite substance, e.g. health is
177
At the end of this passage, Aristotle mentions privation (―ζηέξεζηο‖), but earlier in the passage simply
uses the term absense (―ἀλαηξνπῆο‖). In this passage, he treats the two as equivalent.
84
the substance of disease, for it is by its absence that disease exits;
and health is the formula and the knowledge in the soul.
(1032b2-5)178
I will return to the metaphysical implications of this passage in Section VI below, but for
now, I will examine its epistemological implications. Contraries have the same form
(eidos) and formula (logos). Specifically, they have the form of the positive term. In the
case of health and disease, health is the formula by which one understands disease, in
much the same way that sight is the formula by which one understands blindness.
Therefore, the knowledge (epistēmē) of them is the same.
Curiously, for the rest of Z.7, Aristotle does not again mention privations.
However, he does give a description of the way according to which crafts work and the
relationship between ―thinking and making‖ (―noēsis and poiēsis‖):
Of productions and movements one part is called thinking and
the other making, -that which proceeds from the principle and
the form is thinking, and that which proceeds from the
completion of the thinking is making. (1032b15-17)179
When one understands the principle and the form of something, one ―thinks‖ about it,
Aristotle claims. ―Principle‖ (“archē‖) here implies understanding of causal
connections; it literally means ―origin‖. One can see that this is the case when Aristotle
provides an example from medicine as to the relationship between thinking and making:
―…as in healing, the principle is perhaps the production of warmth, and this the physician
produces by rubbing‖ (1032b25-26).180
Therefore, in order to produce a given result, one
must know two things, the form (eidos) and the origin (archē). Only when one
understands both of these has one completed the thought that will lead to the making.181
However, if this is the case, then Aristotle‘s argument here does not imply a
positive interpretation of the bivalence of crafts. Positive terms and privations have a
great deal in common: form, essence, substance, formula and science. However, they do
not have an archē or origin in common. The physician not only needs to recognize the
178
θαὶ γὰξ ηλ ἐλαληίσλ ηξόπνλ ηηλὰ ηὸ αὐηὸ εἶδνο· ηῆο γὰξ ζηεξήζεσο νὐζία νὐζία ἀληηθεηκέλε, νἷνλ
ὑγίεηα λόζνπ, ἐθείλεο γὰξ ἀπνπζία λόζνο, δὲ ὑγίεηα ὁ ἐλ ηῇ ςπρῇ ιόγνο θαὶ ἐπηζηήκε. 179
Σλ δὴ γελέζεσλ θαὶ θηλήζεσλ κὲλ λόεζηο θαιεῖηαη δὲ πνίεζηο, κὲλ ἀπὸ ηῆο ἀξρῆο θαὶ ηνῦ εἴδνπο
λόεζηο δ‘ἀπὸ ηνῦ ηειεπηαίνπ ηῆο λνήζεσο πνίεζηο. 180
ὥζπεξ θαὶ ἐλ ηῶ ἰαηξεύεηλ ἴζσο ἀπὸ ηνῦ ζεξκαίλεηλ ἀξρή (ηνῦην δὲ πνηεῖ ηῇ ηξίςεη). 181
Note the similarity here between the Phaedrus and the Metaphysics here; one needs to understand both
the whole and causal relationships related to that whole.
85
form of health, but also needs to recognize that heat, in this instance, can produce it.
However, because one needs to understand both the form of something and the ―archē‖
of something in order to produce it, and because knowing the form of something only
provides the form of the privation of that thing and not the ―archē‖ of the privation of
that thing, having the ability to produce something doesn‘t necessarily provide the
positive ability to produce its privation. I will extend Aristotle‘s own example.
Understanding heat (the vibration of particles) provides understanding of coldness (a lack
of that same vibration). However, though I might understand that fire is an origin or
principle of heat, so I can heat something up, this does not imply that I understand
refrigeration. We discovered how to create fire hundreds of thousands of years ago,
while we discovered how to refrigerate only in the mid-Eighteenth Century. Therefore,
Metaphysics Z, too, works best with a negative interpretation of bivalence.
Metaphysics Θ.2 contains Aristotle‘s strongest claims about the bivalence of
crafts, but even this chapter should be interpreted in a negative way.182
This chapter
begins with the claim that crafts are a kind of ―productive knowledge‖ (―poiētikē
epistēmē‖) and are sorts of powers: ―This is why all arts, i.e. all productive forms of
knowledge, are potentialities; they are principles (archai) of change in another thing or in
the artist himself considered as other‖ (1046b3-4).183
This is almost verbatim from the
first sense of ―power‖ in Metaphysics Γ. However, what is most striking about it is his
definition of crafts as ―productive knowledge‖. This definition is unique to this passage
and does not appear in the definition of a craft as a ―reasoned state of capacity to make‖
in Nicomachean Ethics 1140a.184
He refers to crafts as a kind of ―knowledge‖ here, and
does not use the term ―crafts‖ again for the rest of the chapter, referring to crafts
exclusively under the name ―epistēmē‖. He makes the following claim: ―The reason is
that knowledge is a formula, and the same formula explains a thing and its privation…‖
182
For an example of the opposite, positive interpretation, see Makin‘s commentary, in which he holds that
knowledge of the craft of medicine allows one to engage in acts of both healing and contra-healing (as
opposed to just non-healing): ―…and grasp of that single form equips an agent to bring about both Φ-ings
and contra-Φ-ings‖ (Aristotle, Metaphysics: Book [Theta], trans. Stephen Makin [New York: Oxford
University Press, 2006], 49). 183
δηὸ πᾶζαη αἱ ηέρλαη θαὶ αἱ πνηεηηθαὶ ἐπηζηῆκαη δπλάκεηο εἰζίλ· ἀξραὶ γὰξ κεηαβιεηηθαί εἰζηλ ἐλ ἄιιῳ ἠ ᾗ
ἄιιν. 184
ἕμηο ηηο κεηὰ ιόγνπ πνηεηηθή.
86
(1046b8-9).185
The claim that the same formula (“logos‖) explains something and its
privation is familiar from Book Z. So is the claim that something and its privation share
the same knowledge (―epistēmē‖). However, his characterization of what is necessary to
have a craft changes. In Z, someone had a craft by understanding both the form and the
origin. In Θ, one has a craft simply by having the formula. Since a craft is a kind of
knowledge of its product and knowledge is a formula, a craft is a formula of its product.
However, since the formula of a privation is identical to the formula of a positive thing, a
craft is also the formula of its privation. In other words, if all it takes to produce
something is to have its formula, and if the formula of something is identical to that of its
privation, then anyone who is able to produce something must be able to produce its
privation. This might seem to be production in the positive, not the negative sense and
therefore, it would seem to indicate that Aristotle has a positive understanding of
bivalence in Book Θ.
However, any positive interpretation of Aristotle‘s claims in Book Θ would
require having Aristotle argue something unsound. If he is arguing from a positive
interpretation, he would be making the following argument:
1) By understanding something, I understand everything with the same formula.
2) Everything and its privation have the same formula.
3) Therefore, by understanding something‘s formula, I understand its privation‘s
formula.
4) I understand something‘s formula if and only if I am able to produce it.
5) Therefore, if I am able to produce something, I am able to produce its privation.
The problem with this argument on a positive interpretation is that either premise 2) or
premise 4) is false. If I interpret ―formula‖ to exclude what causes something to be
produced, much as ―form‖ and ―origin‖ were separated in Book Z, then premise 4) is
false. Being able only to recognize health is insufficient to produce it. However, if one
includes the ―origins‖ from Book Z in the definition of ―logos‖, then premise 2) becomes
false, because understanding how to produce heat by creating a spark near tinder does not
185
αἴηηνλ δὲ ὅηη ιόγνο ἐζηὶλ ἐπηζηήκε, ὁ δὲ ιόγνο ὁ αὐηὸο δεινῖ ηὸ πξᾶγκα θαὶ ηὴλ ζηέξεζηλ.
87
imply understanding how to cool something by creating a vacuum over ethoxyethane.
So, either understanding something‘s ―formula‖ isn‘t sufficient to constitute a craft or
things and their privations don‘t have the same ―formula‖.
If Aristotle intended there to be a positive interpretation of bivalence in Book Θ,
his argument would be flawed. On the other hand, if the argument is read with a negative
interpretation in mind, the argument would work. On a negative interpretation, one could
take ―formula‖ to include ―archē‖, so that premise 4) is true. Understanding the ―logos‖
of health includes understanding how it is produced and therefore how to produce it.
However, understanding the ―logos‖ of disease includes only knowledge of how to
produce it negatively, not positively. One can ―produce‖ disease in the body by choosing
not to act in the manner that one deliberates would produce health. One can therefore say
that one ―caused‖ the disease through one‘s abstention in the sense in which a ship can be
said to run aground because of the absence of a steersman. The negative interpretation
would therefore not commit Aristotle to any obviously false premises. As long as the
―logos‖ of health includes only how positively to produce health and as long as the
production of disease remains negative, the argument in Book Θ avoids the conflict that
arises between premise 2) and 4) on the positive interpretation.
Aristotle‘s theory of bivalence, then, should be read in a negative way. That is,
Aristotle‘s claims about crafts and their relationship to choice are consistent with an
understanding of capacity for contraries that implies only that a physician can be said to
be responsible for disease in so far as he or she chooses not to heal, rather than implying
that a physician has the positive power to produce disease in a subject who would
otherwise be healthy if not for the physician‘s interference. The most consistent reading
of his position on crafts and contraries in the Metaphysics is that crafts are capable on
contraries in the sense that a rational agent with the power to heal also has the power to
abstain from healing, that is, the physician has a negative power to produce disease.
6 Aristotle’s Solution
Assuming the negative interpretation of bivalence, Aristotle‘s claims about
bivalence are less robust than Plato‘s, who clearly has a positive understanding of
bivalence. Nonetheless, just like Plato, Aristotle will not allow even this negative
88
bivalence to be considered medicine, properly speaking. Like Plato, Aristotle has
linguistic space in which he can carve out the substantive claim that ―per se‖, medicine
cannot be said to be causing disease, even in this negative sense. He makes this claim in
Metaphsysics Θ:
Therefore such [productive] sciences must deal with contraries,
but with one per se and with the other not per se, for the formula
applies to one object per se and to the other in some sense per
accidens. (1046b10-13)186
In this section, I will discuss both the linguistic space in which Aristotle makes this
distinction and his substantive argument for the primacy of producing the positive term.
In the linguistic section, I will discuss per se and per accidens predication, and how such
predication applies to accidents and privations. In the substantive section, I will discuss
Aristotle‘s theory of privation and how privations have a diminished ontological status
relative to their positive terms. Finally, I will look at some of the evidence for why
Aristotle makes the claim that health ought to be considered the positive term while
disease ought to be considered the privation. Together, these various components will
show why Aristotle claims that per se, medicine can only produce health rather than
disease.187
6.1 The Linguistic Space
Just as Plato developed the phrase ―strict speaking‖ in order to argue that a craft,
strictly speaking, can only be used to benefit its subject, Aristotle created a similar
distinction between what he calls predicating ―per se‖ (―kath’auto‖) and ―per accidens‖
(―kata symbebēkos‖) in the Posterior Analytics and the Metaphysics. Using this
186
ὥζη‘ ἀλάγθε θαὶ ηὰο ηνηαύηαο ἐπηζηήκαο εἶλαη κὲλ ηλ ἐλαληίσλ, εἶλαη δὲ ηνῦ κὲλ θαζ‘αὑηὰο ηνῦ δὲ κὴ
θαζ‘ αὑηάο· θαὶ γὰξ ὁ ιόγνο ηνῦ κὲλ θαζ‘ αὑηὸ ηνῦ δὲ ηξόπνλ ηηλὰ θαηὰ ζπκβεβεθόο. 187
Terence Irwin argues for a distinction between ―potentiality‖ (―dynamis‖, which I have been translating
―power‖) and ―possibility‖, and argues that the distinction between causing health and causing disease is a
distinction between the potentiality of medicine and the possibility of medicine. To make this distinction,
Aristotle use bronze, and argues bronze is only possibly a statue, rather than potentially one, because the
archē by which it becomes a statue is external (Metaphysics 1049a13-18). The difficulty with Irwin‘s
interpretation is two-fold: first, Aristotle explicitly argues that crafts, in an accidental sense, are
potentialities for contraries. They may be accidental potentialities, but that does not imply that they are not
potentialities; second, a doctor poisoning and bronze becoming a statue are not analogous on precisely the
relevant point. The physician‘s choice not to heal is the archē of the disease, and this is an internal, rather
than an external archē (Terence Irwin, Aristotle's First Principles [New York: Oxford University Press,
1988], 231-35).
89
distinction, Aristotle carves out the linguistic space he will need to argue that, while
having a craft might make one capable of producing contraries, the term is only used per
se when it is used to create its product, not the privation of that product.
Aristotle uses the term per se predication when something is predicated of
something in terms of what is normally translated ―essence‖, its ―to ti ēn einai‖ or ―to ti
esti‖, literally translated, ―the what it is to be‖ or ―the what it is‖. Aristotle uses the
phrase ―per se‖ to refer to essential predication and ―per accidens‖ to refer to accidental
predication. He makes this distinction in Metaphysics Z: ―The essence (to ti ēn einai) of
each thing is what it is said to be per se‖ (1029b13-14).188
Therefore, per se predication
can be thought of as ―essential‖ predication, in the sense that something is predicated of a
subject per se, when it is a characteristic that makes that thing the kind of thing that it is.
Otherwise, the predication is per accidens. For example, consider a particular human
being, Jon Voight. There are a number of different things that can be predicated of Jon
Voight. For example, one might say that he is ―six-feet tall‖ (a quantity), ―in California‖
(a location), ―an actor‖ (a state), or ―father of Angelina Jolie‖ (a relation). However,
while all of these are characteristics he has, none of these tell you what he is. None of
these describe him as a whole and none of them answer the ―what is Jon Voight?‖
question. The answer to “What is Jon Voight?‖ is ―Jon Voight is a human being‖. The
essence of Jon Voight is his humanity, but ―humanity‖ itself has parts. What makes
something a human being is that it is a rational animal. Being ―rational‖ and being an
―animal‖ are thereby parts of Jon Voight‘s essence; they make Jon Voight the kind of
thing that he is. Therefore being ―rational‖ is not predicated of Jon Voight in the same
way that ―in California‖ is predicated of Jon Voight. Unlike saying ―in California‖,
which relates a non-essential or ―accidental‖ property of Jon Voight, saying ―rational‖
relates an ―essential‖ property of Jon Voight.
This, however, raises a significant issue for the interpretation of Metaphysics Θ.
Crafts and types of knowledge are not essences. One no more says what Hedy Fry is
when one says she is a doctor than one says what Jon Voight is when one says he is an
actor. It would seem, then, that any predication about crafts should be per accidens.
The same is as true of health as it is of crafts. ―Healthy‖ is not a part of the essence of
188
ὅηη ἐζηὶ ηὸ ηί ἤλ εἶλαη ἑθάζηνπ ὃ ιέγεηαη θαζ‘ αὑηό.
90
anything, as one does not become a different type of thing when one becomes ill.
Therefore ―health‖ would appear to be always predicated per accidens as well. If this is
the case, then Aristotle‘s claim that productive knowledge of health is predicated per se
would seem to be nonsense. ―Productive knowledge‖ is never per se and ―health‖ is
never per se, so the term ―per se‖ could apply to neither. If that were the case, then there
would be no way for Aristotle to make a distinction between “per se‖ and ―per accidens‖
predication in crafts because all distinctions would be “per accidens‖
However, Aristotle has another sense of ―essence‖ that I will refer to from this
point on as the ―limited sense‖. In this sense, described in Metaphsyics Z, ―essence‖ can
refer to things other than substances, even to accidents:
But, after all, ‗definition‘, like ‗essence‘, has several meanings;
‗essence‘ (‗to ti estin‘) in one sense means a substance and a
‗this‘, in another one or other of the predicates, quantity, quality
and the like. For as ‗is‘ (‗esti‘) is predicable of all things, not
however in the same sense, but of one sort of thing primarily and
of others in a secondary way, so too the ‗what‘ (‗ti‘)189
belongs
simply to substances, but in a limited sense to the other
categories. (1030a20-25)190
This is a tricky paragraph, because it depends on dividing up the parts of the phrase
normally translated as a single word, ―essence‖: ―to (the) ti (what) esti (it is)‖. What he is
claiming is that ―is‖ has a number of senses, one of which is primary. So, both the
statements, ―Jon Voight is a human‖ and ―Jon Voight is an actor‖ are true, but the ―is‖
here is used differently. The first use which he calls here the ―substantive‖ use is primary
and the second is secondary. Consequently, the term ―what‖ also has primary and
secondary uses. It isn‘t false to answer ―Jon Voight is an actor‖ when one asks what he
is, but one isn‘t giving the primary answer, but rather a secondary one. In fact, Aristotle
is claiming, ―what‖ has just as many senses as ―is‖. For every sense of ―is‖, primary or
secondary, there is a corresponding sense of ―what‖, primary and secondary. What this
implies, though, is that there is also a different sense of essence (the ―what it is‖) that
corresponds to every sense of ―what‖ and ―is‖. This sense of essence is a limited sense of
189
The translation of ―ηη‖ here are ―what‖ rather than ―this‖ is because the term here appears to refer to the
―ηη‖ in ―ηὸ ηί ἐζηηλ‖ above, which is interrogative. 190
ἠ θαὶ ὁ ὁξηζκὸο ὥζπεξ θαὶ ηὸ ηί ἐζηη πιενλαρο ιέγεηαη; θαὶ γὰξ ηὸ ηί ἐζηηλ ἕλα κὲλ ηξόπνλ ζεκαίλεη
ηὴλ νὐζίαλ θαὶ ηὸ ηόδε ηη, ἄιινλ δὲ ἕθαζηνλ ηλ θαηεγνξνπκέλσλ, πνζὸλ πνηὸλ θαὶ ὅζα ἄιια ηνηαῦηα.
ὥζπεξ γὰξ θαὶ ηὸ ἔζηηλ ὑπάξρεη πᾶζηλ, ἀιι‘ νὐρ ὁκνίσο ἀιιὰ ηῶ κὲλ πξώησο ηνῖο δ‘ ἑπνκέλσο, νὕησ θαὶ
ηὸ ηί ἐζηηλ ἁπιο κὲλ ηῇ νὐζίᾳ πὼο δὲ ηνῖο ἄιινηο.
91
essence, and corresponds to the equally limited senses of ―what‖ and ―is‖. In effect, this
means that there are essences of accidents, but in a limited sense.191
This implies that one
can speak per se about accidents. Something is spoken of per se when one attributes to it
something included in its essence; since accidents have essences in a limited sense, one
can make attributions to them per se, when one attributes to it something included in its
limited essence.
Just like substances, accidents have definitions (horismoi).192
Therefore, when
one attributes to an accident something that is a part of the definition of an accident, then
one speaks of it in terms of its limited essence and therefore one speaks of it per se. On
the other hand, when one attributes to an accident something that is not a part of the
definition of that accident, then one speaks of the accident only per accidens. For
example, let us say that I wish to speak about my sight. My sight is not a part of my
essence, as I can be human without having the power of sight. However, I can still speak
about sight per se and per accidens. Sight, like all powers, has a) an effect on b) a
subject, in this case, a) perception b) of colour. Therefore, if I say, ―sight is perception‖,
I am speaking of it per se. ―Perception‖ is a part of the definition of sight, and therefore a
part of its limited essence. However, if I say ―sight is what allows me to read‖, I am
speaking of it per accidens. While this is true, it is not a part of the definition of sight. I
could see before I could read, I would be able to see even if I were illiterate, and many
animals who have no ability to read at all can see. Allowing me to read is not a part of
sight‘s limited essence, but an accidental property.
In this way, Aristotle carves out the linguistic space by which he can say that
medicine is of health ―per se” while it is of disease ―per accidens”. Because ―is‖ has
many different senses clustered around the primary sense of substance, so too do both
―what‖ and ―essence (literally ‗what it is‘)‖. Something is predicated of an accident per
se when it is predicated of it according to the definition of the accident. This still leaves
191
Charlotte Witt argues that Aristotle intends to restrict this limited essence only to its use in per se
predication, and that limited essences don‘t have the same kinds of explanatory roles that substantial
essences have. This is partly correct, as accidents don‘t have the same unity as a substance and therefore
don‘t need an explanation for why they are one thing. However, it would also seem that accidental
essences do have some explanatory role; after all, it is at least in part because of the perception of colour
that I am able to read (Charlotte Witt, Substance and Essence in Aristotle: An Interpretation of Metaphysics
VII-IX [Ithaca: Cornell University Press, 1994], 108-11). 192
This is the same term the Plato used when describing ―criteria‖ of crafts in the Statesman.
92
open the question of on what grounds Aristotle will ascribe healing to medicine per se
and causing disease to medicine per accidens, but in order to understand this position, we
must turn instead to Aristotle‘s substantive argument.
6.2 The Substantive Argument
Aristotle‘s distinction between per se and per accidens predication provides him
with the linguistic distinction he needs in order to make the claim that medicine‘s power
to heal is in some way primary compared to its power to harm. However, I have not yet
provided an argument as to why one ought to consider healing medicine per se but
causing disease medicine per accidens. In order to understand this, one must look at his
ontological distinctions between a positive state and a privation and his arguments for
why a privation can never be predicated of anything per se. Because privations can never
be predicated of anything per se and because disease is a privation, disease can never be
predicated of anything per se and there is no knowledge of disease per se, as opposed to
health of which there is both per se predication and per se knowledge. As a result, both
disease and knowledge of disease are always predicated accidentally and they are not an
essential part of anything, whether in the full or limited sense of essential, including the
craft of medicine.
I will return now to Metaphysics Z to investigate how Aristotle uses privation in
his account of why disease can only be predicated per accidens. Though I have quoted
this passage before, I will repeat an extended version of the passage here for reference:
… from art proceed the things of which the form is in the soul.
(By form I mean the essence of each thing and its primary
substance.) For even contraries have in a sense the same form;
for the substance of a privation is the opposite substance, e.g.
health is the substance of disease; for it is by its absence that
disease exists; and health is the formula and the knowledge in
the soul. (1032b2-5) 193
This is an important passage and in many ways a puzzling one. Aristotle is claiming that
quite a few things are the same between something and its privation: the form, the
193
ἀπὸ ηέρλεο δὲ γίγλεηαη ὅζσλ ηὸ εἶδνο ἐλ ηῇ ςπρῇ (εἶδνο δὲ ιέγσ ηὸ ηί ἤλ εἶλαη ἑθάζηνπ θαὶ ηὴλ πξώηελ
νὐζίαλ)· θαὶ γὰξ ηλ ἐλαληίσλ ηξόπνλ ηηλὰ ηὸ αὐηὸ εἶδνο· ηῆο γὰξ ζηεξήζεσο νὐζία νὐζία
ἀληηθεηκέλε, νἷνλ ὑγίεηα λόζνπ, ἐθείλεο γὰξ ἀπνπζία λόζνο, δὲ ὑγίεηα ὁ ἐλ ηῇ ςπρῇ ιόγνο θαὶ
ἐπηζηήκε.
93
essence, the ―primary substance‖, the substance and finally the formula or ―logos‖. This
passage is a virtual survey of Aristotelian metaphysical principles, and Aristotle identifies
all of them in the case of something and its privation. This identification provides the
basis of his argument that crafts are capable of contraries in the last section. However,
this passage also serves as the beginning of an argument subordinating privations in
important ways such that any knowledge of privations is accidental and privations can
never be predicated per se.194
In Book Θ, Aristotle returns to the subject of contraries and demotes the
knowledge of the privation to something that is only accidentally the subject of a science
and not essentially. This is quite a long passage, but I will quote it in full as I will be
referring to it a great deal:
The reason is that science is a rational formula, and the same
rational formula explains a thing and its privation, only not in the
same way; and in a sense it applies to both, but in a sense it
applies rather to the positive fact. Therefore such sciences must
deal with contraries, but with one per se and with the other not
per se; for the rational formula applies to one object per se, and
to the other, in a sense, per accidens. For it is by denial and
removal that it explains the contrary; for the contrary is the
primary privation, and this is the entire removal of the positive
term. (1046b8-15) 195
Therefore, despite the earlier claims in Book Z that a thing and its contrary share the
same form, essence, substance and formula, Aristotle explicitly demotes privations in
Book Θ to the position of something known only accidentally. Somehow, privations are
not the sort of thing that can be an essential part of any body of knowledge.
194
Note that Aristotle claims in the above passage that all contraries have a single form. This means that
the apparent continuum from heat to cold, for example, must be defined relative to one of the extremes, as
our later discovery that heat is a motion of particles would confirm. In The concept of first philosophy and
the unity of the Metaphysics of Aristotle, Giovanni Reale and John R. Catan make this same observation
(Giovanni Reale and John R. Catan, The Concept of First Philosophy and the Unity of the Metaphysics of
Aristotle [Albany: State University of New York Press, 1980], 264). 195
αἴηηνλ δὲ ὅηη ιόγνο ἐζηὶλ ἐπηζηήκε, ὁ δὲ ιόγνο ὁ αὐηὸο δεινῖ ηὸ πξᾶγκα θαὶ ηὴλ ζηέξεζηλ, πιὴλ νὐρ
ὡζαύησο, θαὶ ἔζηηλ ὡο ἀκθνῖλ ἔζηη δ‘ ὡο ηνῦ ὑπάξρνληνο κᾶιινλ, ὥζη‘ ἀλάγθε θαὶ ηὰο ηνηαύηαο ἐπηζηήκαο
εἶλαη κὲλ ηλ ἐλαληίσλ, εἶλαη δὲ ηνῦ κὲλ θαζ‘αὑηὰο ηνῦ δὲ κὴ θαζ‘ αὑηάο· θαὶ γὰξ ὁ ιόγνο ηνῦ κὲλ θαζ‘
αὑηὸ ηνῦ δὲ ηξόπνλ ηηλὰ θαηὰ ζπκβεβεθόο· ἀπνθάζεη γὰξ θαὶ ἀπνθνξᾷ δεινῖ ηὸ ἐλαληίνλ· γὰξ ζηέξεζηο
πξώηε ηὸ ἐλαληίνλ, αὕηε δὲ ἀπνθνξὰ ζαηέξνπ.
94
Aristotle provides definitions of privations and contraries in Metaphysics Γ.
There, he gives two senses of privation that are especially important to understanding
why privations are only predicated per accidens. Those first two are the following:
We speak of privation (1) if something has not one of the
attributes which a thing might naturally have, even if this thing
itself would not naturally have it, e.g. a plant is said to be
deprived of eyes. (2) If, though either the thing itself or its genus
would naturally have an attribute, it has it not, e.g. a blind man
and a mole are in different senses deprived of sight; the latter in
contrast with its genus, the former per se. (1022b23-27)196
Note that these two senses of privation are importantly distinct. We say that plants are
sightless, but we do not say that plants are blind. The first sense of privation is the sense
in which anything can be said to have a privation if it lacks that property, whether it is
normal for that thing to have the property or not. So, for instance, I am ―flightless‖ in
this sense and I also lack an exoskeleton in this sense. In the definition, he limits these
kinds of privation to properties that he says that ―a thing might naturally have‖. This is
an incredibly broad category and would seem to include things like ―has a thousand
eyes‖, as a fly has. In fact, he expands this category even further a paragraph later when
he says, ―There are just as many kinds of privations as there are of words with negative
prefixes‖ (1022b33).197
The second sense of privation is privation of a property that is natural to an
individual‘s kind. So, for example, a human being is blind rather than simply sightless
because human beings are naturally sighted. Aristotle‘s example of the mole is quite
interesting. Qua animal, a mole is blind, because animals are naturally sighted; however,
qua mole, a mole is merely sightless, because moles are not naturally sighted. It is in this
second sense that a person would be said to be sick. People are normally healthy;
therefore, sickness is a privation in this second sense. Only an organism can be sick;
rocks can merely be non-healthy. Therefore, when something lacks some characteristic
that it would normally have in virtue of its nature, it has a privation in this second sense.
Note that privations of the second sort are also privations in the first sense. Blind people
196
ηέξεζηο ιέγεηαη ἕλα κὲλ ηξόπνλ ἂλ κὴ ἔρῃ ηη ηλ πεθπθόησλ ἔρεζζαη, θἂλ κὴ αὐηὸ ᾖ πεθπθὸο ἔρεηλ,
νἷνλ θπηὸλ ὀκκάησλ ἐζηεξῆζζαη ιέγεηαη· ἕλα δὲ ἂλ πεθπθὸο ἔρεηλ, ἠ αὐηὸ ἠ ηὸ γέλνο, κὴ ἔρῃ, νἷνλ ἄιισο
ἄλζξσπνο ὁ ηπθιὸο ὄςεσο ἐζηέξεηαη θαὶ ἀζπάιαμ, ηὸ κὲλ θαηὰ ηὸ γέλνο ηὸ δὲ θαζ‘ αὑηό. 197
αἱ ἀπὸ ηνῦ α ἀπνθάζεηο ιέγνληαη, ηνζαπηαρο θαὶ αἱ ζηεξήζεηο ιέγνληαη.
95
are also sightless and sick people are also non-healthy, but not vice versa. Aristotle also
distinguishes a third and fourth sense that I will not use but should be mentioned. The
third is a mixed use, according to which one says that something of a certain age has a
privation in the first sense even when it is natural to a member of its species in the second
sense. For example, children are not fertile, but are not infertile. Although fertility is
natural to human beings, it is not natural to human beings of that age, so one speaks of
their privation as though it is of the first sense rather than the second. The fourth sense is
a homonymous sense of someone who has been forcibly robbed or ―deprived‖ of
something that was theirs that, like most homonyms, does not translate well into English.
All privations, that is, all negative predications, are included in the first sense of
privation, while a smaller set, negative predications of things that would naturally be
otherwise, are of the second sense of privation. Disease is a privation of health of this
second sense as human beings are naturally healthy.198
With these definitions of ―privation‖ in mind, I will return to the passages from
Metaphysics Z and Θ. One will notice that, even when Aristotle is claiming that things
and their privations have the same form, essence, substance and formula, the relationship
between them is still asymmetrical. His example is that, ―Health is the substance of
disease‖ (1032b4). Notice, though, that he does not say that disease is the substance of
health. Rather, health and disease both have the same substance: that of health. This is
the case with any positive term and its privation: the positive term and the privation have
the same essence, that of the positive term. The asymmetry of this relationship does not
imply that terms and their privations are equivalent because they share a form, essence,
substance and formula, but rather that the privation is entirely dependent on the positive
term for its form, essence, substance and formula. One should therefore not read the
claims in Book Z that things and privations have the same form, essence, substance and
formula as in any way equating them. Instead, Book Z introduces the subordination of
privations to their positive terms.
198
Though Aristotle does not mention this explicitly, it is common for privations of the second kind to have
independent terms that are not mere negations, like ―blind‖ as opposed to ―sightless‖. ―Sick‖ or ―nosos‖ is
such a word in Greek, as it is in English. English (unlike Greek) also denotes second-type privations by
using the ―in-― prefix instead of a ―non-― or ―a-― prefix.
96
Metaphysics Θ makes this subordination far more explicit, though it has the effect
of making it in some ways more perplexing. First, Aristotle speaks about the formula or
―logos‖ that a thing and its positive term share and says that this logos applies to both, but
not to them both in the same way. Specifically, he says that the formula applies to the
thing per se, while it applies to the privation per accidens. Aristotle thereby introduces
the per se and per accidens language that I discussed above in the ―Linguistic Space‖
section. However, one must still examine why it is that privations are only said of
something per accidens rather than per se. Aristotle claims that the formula is predicated
of the privation accidentally because it explains the privation through ―denial and
removal‖. Somehow, because the formula or a term explains privation by denial and
removal, it only explains that privation per accidens. We therefore have the following
series of claims:
1) A formula (logos) explains a thing (pragma) and its privation (sterēsis).
2) It explains the thing per se and the privation per accidens.
3) It explains the privation per accidens because it explains the privation through
denial (apophasis) and removal (apophōra)
We are told that the formula explains both things and their privations, and then told that it
only explains the latter per accidens because it explains it through denial and removal.
While this is an argument, the connection between the various premises are not
immediately obvious. To understand why one should be believe that explanations
through denial and removal should be considered per accidens explanations, we must
turn to other texts. Two of the three uses of ―apophora‖ in the Aristotelian corpus appear
in the above passage (the other concerns farmers holding property in the Politics 1264a),
so there is little extant evidence of what he may have meant by the term. However, he
has a great deal to say about ―denial‖ and why ―denial‖ is always accidental, especially in
the Topics and the Parts of Animals, so I will turn to those.
In the Topics, Aristotle criticizes those who would divide a genus using a
privation, such as the division of line into those with breadth and those without breadth:
―Moreover, see if he divides the genus by a negation, as those do who define a line as
length without breadth; for this means simply that is has not any breadth‖ (143b11-
97
12).199,200
The difficulty with dividing a genus by a privation, he says, is that it would
imply that the genus was a member of its own privative species. If one was to divide
lines by those with breadth and those with no breadth, one would end up with the
following tree:
The problem with this tree is that the genus, lines, does not include the property of
breadth. The genus and the non-breadth species, therefore, would collapse into each
other as there would be nothing to differentiate the genus from the species: ―The genus
will then be found to partake of its own species‖ (143b14).201
Therefore, a privation can
never serve as a differentia of anything, because it would collapse genera and species
together.202
The reason given in the Parts of Animals is different than this. The problem
presented there is that, if a privation were used as a differentia, it would not itself admit
of any subdivision: ―But privative terms in their character of privatives admit of no
subdivision‖ (642b23-24).203
For instance, footed may be divided into hoofed and soft-
footed. However, there is no way to subdivide being footless. So, for instance, if one
was to try to divide using privative differentia, one would be forever stuck with
indivisible privations:
199
Ἔηη ἐὰλ ἀπνθάζεη δηαηξῇ ηὸ γέλνο, θαζάπεξ νἱ ηὴλ γξακκὴλ ὁξηδόκελνη κῆθνο ἀπιαηὲο εἶλαη· νὐδὲλ
γὰξ ἄιιν ζεκαίλεη ἠ ὅηη νὐθ ἔρεη πιάηνο. 200
Plato also argues for this position, though with a different sort of argument, in Statesman 262c-263a,
where he argues that one should not separate Greeks off from ―barbarians‖ because such a division does not
separate according to real classes and does not divide properly in two. 201
ζπκβήζεηαη νὖλ ηὸ γέλνο κεηέρεηλ ηνῦ εἴδνπο. 202
David M. Balme argues that this argument is directed specifically against the Platonists, who argue that
forms are things. Either the form of line has breadth or it does not. If it does not, then it is identical to the
form of ―lines-without-breadth‖ (Aristotle, De Partibus Animalium I and De Generatione Animalium I, ed.
David M. Balme [New York: Oxford University Press, 1992], 109). 203
Οὐθ ἔζηη δὲ δηαθνξὰ ζηεξήζεσο ᾗ ζηέξεζηο· ἀδύλαηνλ γὰξ εἴδε εἶλαη ηνῦ κὴ ὄληνο.
Lines
Lines with Breadth
Lines without Breadth
98
This results from Aristotle‘s differentia being importantly adverbial. They separate
different ways of being a member of the genus, not simply different subsets.204
This is
why non-footed animals couldn‘t simply be divided into red non-footed animals and blue
non-footed animals, because red and blue are not different ways of being non-footed,
while being hoofed and soft-footed are different ways of being footed.205
As a result, no
privation can ever serve as a differentia, because no privation could itself take differentia.
What these two examples from the Topics and The Parts of Animals show is that
privations can never be a part of the definition of anything, because they can never serve
as a defining or essential characteristic of anything. Privations are simply incapable of
being a part of any definition, because they can never be used to divide one sort of thing
from another sort of thing (or even any sort of accident from another sort of accident,
since accidents have species and differentia of their own). Since per se predication
requires that what is predicated be a part of a definition and since a privation can never be
a part of any definition, any privation is necessarily predicated per accidens, rather than
per se.
204
Allan Bäck takes this claim one step further, arguing that Aristotle believes even that predicates are
adverbial, that is, they are ways of existing: ―...my characterization of a tripartite sentence, ‗S is P‘, as
taking the term ‗P‘ as an adverbial qualification of the statement of existence: ‗S is, as a P.‘‖ Allan T. Bäck,
Aristotle's Theory of Predication (Boston: Brill, 2000), 209. 205
James G. Lennox provides a discussion of how differentiae should be based on different ―ways‖ of
being the genus of which it demarcates a species. He further argues that Aristotle is consciously criticizing
the kind of division that occurred in the Sophist and Statesman, where Plato does use negative predication
(despite his own argument against the term ―barbarian‖), such as dividing footed animals into horned and
hornless at 265c (Aristotle, On the Parts of Animals I-IV, trans. James G. Lennox [New York: Oxford
University Press, 2001], 165-66).
Animal
Footed
Hoofed Soft
Non-footed
? ?
99
6.3 Aristotle on Disease as a Privation
So far, I have discussed why it is that Aristotle believes that crafts should not be
said to produce privations per se. However, I have not yet addressed Aristotle‘s
arguments that disease is, indeed, a privation. Why, for example, should disease not be
an opposite with its own form, rather than one that is simply a privation of the form of
health? Aristotle doesn‘t provide the same sort of detailed metaphysical analysis of this
question that Plato does in the Philebus. However, Aristotle does say a number of things
about health that would imply that his beliefs about it are quite similar to Plato‘s.
Specifically, health is treated as a kind of balance in the body. The opposite of balance is
not some sort of opposing extreme, but rather, simply its absence, imbalance. In this
section, I will collect the evidence for Aristotle‘s claim that disease is a privation of
health and demonstrate that similarity with Plato.
First, I‘ll give a few words on the definition of health. Aristotle equates health
with ―the‖ or sometimes ―a‖ good condition (hexis) of the body in several places, such as
De Anima I ―one of the excellences of the body‖ (408a1),206
Eudemian Ethics II ―the best
condition of the body‖ (1220a19-20),207
and the Rhetoric ―The excellence of the body is
health‖ (1361b3-4).208
The definition here appears to be analytic. No argument about
whether or not health is good for the body is necessary, because health is by definition
what is good for the body. This claim appears in the Eudemian Ethics I, where Aristotle
says that no one would even try to prove that health is a good, as that is a principle:
Further, no one demonstrates that health is good (unless he is a
sophist and no doctor, but one who produces deceptive
arguments from inappropriate consideration), any more than any
other principle. (1218b21-24)209
According to Aristotle, then, health is the good condition of the body, which is good by
definition. In other words, whatever constitutes the good condition of the body is what
Aristotle calls health.
206
ηλ ζσκαηηθλ ἀξεηλ. 207
ὅηη ἀξίζηε δηάζεζηο ηνῦ ζώκαηνο. 208
ζώκαηνο δὲ ἀξεηὴ ὑγίεηα. 209
ἔηη νὐδὲ δείθλπζηλ νὐζεὶο ὅηη ἀγαζὸλ ὑγίεηα, ἂλ κὴ ζνθηζηὴο ᾖ θαὶ κὴ ἰαηξόο (νὗηνη γὰξ ηνῖο
ἀιινηξίνηο ιόγνηο ζνθίδνληαη), ὥζπεξ νὐδ‘ ἄιιελ ἀξρὴλ νὐδεκίαλ.
100
However, Aristotle isn‘t content with simply this analytic definition and provides
health with some positive characteristics as well. Specifically, health is described as a
balance, a harmony and a mean between different types of opposites. For example, in De
Anima I, Aristotle makes the following Platonic remarks:
Harmony, however, is a certain proportion or composition of the
constituents blended and the body is composed of contraries…It
is more appropriate to call health (or generally one of the good
states of the body) a harmony than to predicate it of the soul.
(407b31-32,408a1-2) 210
This passage argues against the claim that the soul is a harmony in very similar terms to
that used in the Phaedo, and makes the claim that health is a harmony of various
opposites in the body. This theme of appropriate blending appears starkly in the Physics
VII:
Further, we say that all excellences depend upon particular
relations. Thus bodily excellences such as health and fitness we
regard as consisting in a blending of hot and cold elements in
due proportion, in relation either to one another within the body
or to the surrounding. (246b3-6) 211
In Parts of Animals, Aristotle adds dry and moist to this list:
There ought, then, to be some clear understanding as to the sense
in which natural substances are to be termed hot or cold, dry or
moist. For it appears manifest that these are properties on which
even life and death are largely dependent, and that they are
moreover the causes of sleep and waking, of maturity and old
age, of health and disease… (648b2-6)212
Both of these texts show Aristotle views health in much the same way as Plato did, that
is, as a kind of harmony between the various contraries in the body. Aristotle specifies
these contraries in a way that Plato does not, referring to heat and cold in the Physics and
adding dry and moist in the Parts of Animals.
210
θαὶ γὰξ ηὴλ ἁξκνλίαλ θξᾶζηλ θαὶ ζύλζεζηλ ἐλαληίσλ εἶλαη, θαὶ ηὸ ζκα ζπγθεῖζζαη ἐμ
ἐλαληίσλ…ἁξκόδεη δὲ κᾶιινλ θαζ‘ ὑγηείαο ιέγεηλ ἁξκνλίαλ, θαὶ ὅισο ηλ ζσκαηηθλ ἀξεηλ, ἠ θαηὰ
ςπρῆο. 211
ἔηη δὲ θαί θακελ ἁπάζαο εἶλαη ηὰο ἀξεηὰο ἐλ ηῶ πξόο ηη πὼο ἔρεηλ. ηὰο κὲλ γὰξ ηνῦ ζώκαηνο, νἷνλ
ὑγίεηαλ θαὶ εὐεμίαλ, ἐλ θξάζεη θαὶ ζπκκεηξίᾳ ζεξκλ θαὶ ςπρξλ ηίζεκελ, ἠ αὐηλ πξὸο αὑηὰ ηλ ἐληὸο ἠ
πξὸο ηὸ πεξηέρνλ. 212
Γηὸ δεῖ κὴ ιαλζάλεηλ πο δεῖ ηλ θύζεη ζπλεζηώησλ ηὰ κὲλ ζεξκὰ ιέγεηλ ηὰ δὲ ςπρξὰ θαὶ ηὰ κὲλ μεξὰ
ηὰ δ‘ ὑγξά, ἐπεὶ ὅηη γ‘ αἴηηα ηαῦηα ζρεδὸλ θαὶ ζαλάηνπ θαὶ δσῆο ἔνηθελ εἶλαη θαλεξόλ, ἔηη δ‘ ὕπλνπ θαὶ
ἐγξεγόξζεσο θαὶ ἀθκῆο θαὶ γήξσο θαὶ λόζνπ θαὶ ὑγηείαο.
101
Treating health as an appropriate balance of opposites allows Aristotle to treat
health as though it has a determinate account, while disease does not. Similarly to Plato
in the Philebus, Aristotle believed that disease is indeterminate, that is, there are any
number of ways that something can deviate from health, but only one way that something
can be considered healthy.213
For instance, in the Topics, Aristotle makes the observation
that, while health has a contrary in disease, the various diseases do not have contraries:
―…for health without question is the contrary of disease, whereas a particular disease,
e.g. fever and ophthomalmia and any other particular disease, has no contrary‖ (123b33-
36).214
Disease has a multiplicity to it that means, while one can contrast ―health‖ and
―disease‖, particular diseases themselves have no contraries. Moreover, in the
Nicomachean Ethics, Aristotle compares health to the virtues, which themselves require a
mean:
It is evident too that both are bad, being unjustly treated and
acting unjustly; for the one means having less and the other
having more than the intermediate amount, which plays the part
here that the healthy does in the medical art… (1138a29-31)215
On the other hand, health is singular and determinate. For example, Aristotle claims in
Metaphysics Γ that, ―As, then, there is one science which deals with all healthy things,
the same applies in the other cases also‖ (1003b11-12).216
Disease is so multiform because it is a corruption of a natural state. Metaphysics
H makes a distinction between something‘s natural state and a corruption of that natural
state:
E.g. if the body is potentially healthy, and disease is contrary to
health, is it potentially both?...We answer that it is the matter of
one in virtue of its positive state and its form, and of the other in
213
Aristotle‘s language is somewhat different from Plato‘s. Plato says that disease is ―undefined‖
(―apeirōn‖), a privative phrase, while Aristotle says that it is ―multiform‖ (―polueides‖). Both terms,
however, imply that disease lacks a single, intelligible form. 214
(ἔλζηαζηο ἐπὶ ηῆο ὑγηείαο θαὶ λόζνπ·ἁπιο κὲλ γὰξ ὑγίεηα λόζῳ ἐλαληίνλ, δὲ ηὶο λόζνο εἶδνο νὖζα
λόζνπ νὐδελὶ ἐλαληίνλ, νἷνλ ὁ ππξεηὸο θαὶ ὀθζαικία θαὶ ηλ ἄιισλ ἕθαζηνλ.) 215
θαλεξὸλ δὲ θαὶ ὅηη ἄκθσ κὲλ θαῦια, θαὶ ηὸ ἀδηθεῖζζαη θαὶ ηὸ ἀδηθεῖλ (ηὸ κὲλ γὰξ ἔιαηηνλ ηὸ δὲ πιένλ
ἔρεηλ ἐζηὶ ηνῦ κέζνπ θαὶ ὥζπεξ ὑγηεηλὸλ κὲλ ἐλ ἰαηξηθῇ. 216
θαζάπεξ νὖλ θαὶ ηλ ὑγηεηλλ ἁπάλησλ κία ἐπηζηήκε ἔζηηλ, ὁκνίσο ηνῦην θαὶ ἐπὶ ηλ ἄιισλ.
102
virtue of the privation of the positive state and the corruption of
it contrary to its nature. (1044b30-34)217
Disease is also described as such a corruption of health in De Longitudine et Brevitate
Vitae, in which Aristotle says, ―…in many things their mode of corruption is something
peculiar to themselves, e.g. in knowledge and ignorance, in health and disease‖ (465a19-
21).218
Disease is described as more than just a privation of health here, but as the
corruption (―phthora”) of health, a term that can literally mean ―destruction‖.
When one takes together all of these claims about health, why Aristotle would
claim that health is the positive term and disease is the privation becomes obvious.
Health is, by definition, the good state of the body, which is defined, like all good states,
with reference to the nature of the body. Therefore, anything that deviates from this good
state will be considered a disease. Diseases, therefore, are indeterminate, as there are
many different ways to deviate from something, and so disease is multiform and
undefined except as an absence of the positive state. Aristotle‘s particular theory of
health is that it is the result of or perhaps consists in a harmony within the body of heat
and cold, and dry and wet. As with any harmony, there are multiple ways to deviate from
a given harmony whose only characteristic in common is that they deviate from the same
harmony. Considered together, then, Aristotle‘s various comments about disease show
why he would claim that health is the positive state, while disease is the privation. Health
is the good condition of the body, while disease is any deviation from (i.e. privation of)
that good condition.
6.4 Conclusion
Like Plato, Aristotle believes that, in some sense, a craft is capable of contraries,
but that one can only be said to be applying a craft in a strict sense (per se) when one
applies it in order to produce the positive fact. In this section, I discussed his argument,
first discussing whether he meant this in the positive way that a physician can create
disease in a healthy patient or in the negative way that a physician can ―create‖ (be
217
. νἷνλ εἰ ηὸ ζκα δπλάκεη ὑγηεηλόλ, ἐλαληίνλ δὲ λόζνο ὑγηείᾳ, ἆξα ἄκθσ δπλάκεη; θαὶ ηὸ ὕδσξ δπλάκεη
νἶλνο θαὶ ὄμνο; ἠ ηνῦ κὲλ θαζ‘ ἕμηλ θαὶ θαηὰ ηὸ εἶδνο ὕιε, ηνῦ δὲ θαηὰ ζηέξεζηλ θαὶ θζνξὰλ ηὴλ παξὰ
θύζηλ; 218
εἰζὶ γὰξ ἴδηαη θζνξαὶ πνιινῖο ηλ ὄλησλ, νἷνλ ἐπηζηήκῃ <θαὶ ἀγλνίᾳ>, θαὶ ὑγηείᾳ θαὶ λόζῳ.
103
causally responsible for) disease through abstention from treatment. I concluded that
Aristotle‘s argument is best understood using a negative interpretation. Next, I discussed
the use of per se predication as opposed to per accidens predication, specifically with
respect to the per se predication of accidents, that is, attribution of accidents according to
their definition. I then established that, since privations can never be predicated per se,
medicine can only be said to be used per se when used for the positive term. I then
considered why it was that Aristotle believed that health was a positive term and disease
a privation, showing that, because health was by definition the positive condition of the
body, it is uniform while deviations from it are multiform, and considered the evidence
from the Aristotelian corpus to that effect.
7 Chapter Conclusion
Plato and Aristotle both hold two positions that appear largely in conflict with
each other. On the one hand, they both hold that crafts are capable of contraries. For
Plato, this is clearly in a positive sense, in which having a craft enables one to perform
the exact opposite task, stealing as well as guarding or poisoning as well as healing. For
Aristotle, this is best interpreted as a negative position, in which he claims that having a
power gives one the choice one might not otherwise have had to not perform the task
associated with the craft. On the other hand, however, both wish to claim that, strictly
speaking, such non-uses of the crafts are not a part of the craft, and both carve out both
linguistic space in which to provide a substantive argument. Plato develops a sense of
strict speaking, according to which a craft is only strictly speaking applied when it is does
with knowledge and for the benefit of the patient. Aristotle separates per se from per
accidens predication, the former of which refers to all speech that refers to the essential
properties of something and the latter of which refers to all other speech. Substantively,
Plato equates cause with the fourth kind of thing in the Philebus, that which generates
order out of chaos using measurement, which can only be beneficial for its subject.
Aristotle in turn demotes privations to a purely derivative status, dependent on positive
terms for their meanings. Together, both of them hold both the view that being a doctor
gives one the power to poison, but that poisoning is not strictly speaking a part of
medicine.
104
C h a p t e r 3 : P l a t o o n P e r s u a s i o n a n d K n o w l e d g e
1 Introduction
In this chapter, I will discuss Plato‘s argument for knowledge as the proper
criterion for freedom and show how he applies this criterion in a medical context. Plato
develops this conception as the response to the disintegration of what I will call the
―traditional synthesis‖. Traditionally, as I will show, Athenians made some quite clear
associations. On the one hand, freedom and persuasion were allied while, on the other
hand, force and slavery were allied. Democratic Athens was considered free because
their political system operated by persuasion, while others, especially the Asians, were
slaves because their political systems operated by force. Even before Plato‘s birth,
however, the traditional synthesis was breaking down. A new group of professional
rhetors, especially the Sicilian Gorgias, threatened this traditional synthesis. They
claimed that they were able to teach persuasion so well as to make persuasion a kind of
force, and they promised young men that they could teach them to enslave others through
rhetoric. This threatened the traditional synthesis.
In response to this disintegration, Plato does not simply defend the traditional
synthesis as does, for example, Lysias. Instead, Plato develops a new conception of
freedom in response to Gorgias‘ challenge. On the one hand, he accepts Gorgias‘ claim
that persuasion, at least non-rational persuasion, is no ally of freedom. However, this
does not mean that freedom is an illusion. Rather, knowledge is the true ally of freedom.
Plato delimits two different types of persuasion: sorcerous and didactic persuasion.
Didactic persuasion can provide freedom for the hearer by being an antidote to the
sorcerous persuasion of Gorgias and by being the appropriate sort of learning for a free
person. I will conclude the chapter with a look at Plato‘s discussion of informed consent
in the Laws and, given the evidence available, discuss just what kind of knowledge Plato
might have envisioned as being appropriate to a free person in a medical context.
105
2 The Texts
I will use several texts in this chapter, and I will here provide a short summary of
them. The first group of texts will be used as demonstrations of the traditional synthesis,
while the next group will provide Gorgias‘s challenge and Plato‘s new characterization of
freedom. I will refer to other texts as well, but the texts listed below will receive the most
attention:
Group 1:
Suppliant Maidens by Aeschylus (466-455 B.C.): 219
The daughters of Danaos flee forced
marriages to the Egyptians and seek the refuge of the Argives under king Pelasgos. He
grants them asylum on the grounds that persuasion rather than force is an ally of freedom.
Heraclidae by Euripides (430 B.C.): Eurystheus wishes to kill the children of Heracles,
whom he pursues to Athens. Demophon agrees to protect the children and a debate
ensues about whether Demophon should return the children. Ultimately, it is decided that
a free people must be persuaded rather than threatened.
Philoctetes by Sophocles (409 B.C.): Neoptolemus and Odysseus go to the island of
Lemnos in order to retrieve the bow of Heracles from Philoctetes. In their debates with
each other and with Philoctetes about the noble and ignoble ways of retrieving the bow,
they consider the relationships between force, persuasion, trickery, freedom and class.
The Funeral Oration by Lysias (395-387 B.C.): 220
In a speech in praise of the Athenian
dead in the Corinthian War, Lysias provides perhaps the clearest statement of the
traditional synthesis.
219
Composition dates for the tragedians are taken from Simon Hornblower and Anthony Spawforth, The
Oxford Companion to Classical Civilization (Oxford: Oxford University Press, 2004). 220
This speech was written at an unknown time during the war between the Corinthians and the Athenians,
which lasted from 395-387 B.C. The speech was given during an annual feast in honour of those who died
during the previous year, the same feast in which Pericles delivered his famous funeral oration.
(Thucydides, The History of the Peloponnesian War, Books 1-2, trans. C. F. Smith [London: Heinemann,
106
Group 2:
The Encomium of Helen by Gorgias (5th
century B.C.): 221
In this showcase speech,
Gorgias defends Helen on the grounds that, if she was forced by Paris to go to Troy,
Helen was not responsible, but if she was persuaded, it was still a sort of force.
Gorgias 447-465 by Plato (early-mid 4th
century B.C.): In this dialogue, Socrates debates
with Gorgias about rhetoric, raising concerns about freedom, persuasion and competition.
Lysis, esp. 207-210 by Plato (early-mid 4th
century B.C.): In this dialogue, Socrates
discusses knowledge and its relationship to freedom, especially in the area of craft
knowledge.
Republic VII by Plato (early-mid 4th
century B.C.): In this book, Socrates discusses the
various ways in which freedom and persuasion interact, largely using the metaphor of
sorcery.
Laws 720a-d, 857c-d by Plato (mid 4th
century B.C.): At two separate points in the Laws,
the Athenian visitor explicitly discusses the relationship between persuasion, knowledge
and freedom in a medical context. He draws contrasts between free and slave doctors and
patients, and the appropriate ways to deal with each.
On the Art by ―Hippocrates‖ (5th
-4th
century B.C.): One of the authors of the Hippocratic
corpus discusses the difficulties that Hippocratic doctors have promoting their craft in a
public space.
1919, 2.34-46) and which served as the forum for the oddly inaccurate Platonic dialogue Menexenus (S. C.
Todd, A Commentary on Lysias: Speeches 1-11 [Oxford: Oxford University Press, 2007], 149-53). 221
Gorgias lived 90 years and there is no evidence as to when the Encomium was written. It could even
have been written as late as 387 B.C., the date of his death.
107
3 Eleutheria, Freedom and Autonomy
The central term that will be discussed in this chapter, ―eleutheria‖, requires some
preliminary discussion because it has a similar but not identical semantic field to its usual
English equivalent, ―freedom‖. According to the Greek-English Lexicon, the word
“eleutheros‖ has a number of meanings, including the related meanings ―freedom‖,
―independent‖, ―open to all‖, ―unencumbered by‖, and the related meanings ―fit for a
freeman‖, and ―frank‖.222
Its opposite is ―doulos‖ or ―slave‖. For all of these uses, there
is an analogous use for the English word ―free‖ in the Oxford English Dictionary.223
―Freedom‖ is obvious. ―Independent‖ appears in an example for definition 6.a.: ―Since I
was what may be termed a free man; or, in other words, since I became independent.‖
―Unencumbered‖ appears in an example under the same heading: ―Finally the divorce
mills got through their slow grinding and both the daughter and I were free and
unencumbered‖. ―Open to all‖ is part of definition 9.b.: ―b. Open to all competitors;
open for all. free fight: a fight in which all present may join.‖. ―Frank‖ is a part of
definition 23: ―23. Frank and open in conversation or intercourse, ingenuous,
unreserved‖. The same contrast with slavery also exists in the first definition of the term
―freeman‖: ―1.a. One who is personally free; one who is not a slave or serf‖.
Two things should be noticed here concerning these definitions that will be
important to understanding their use. First, the term ―eleutheria‖ is robust. In modern
ethics, much of the ethical work that ―freedom‖ once did has been taken up by the term
―autonomy‖, especially in the area of medical ethics. Second, the term serves as a class
term. In a society with slavery, like ancient Greece, being free rather than a slave marked
one as a member of a social class. Certain behaviours might be appropriate or
inappropriate to a free person as a member of their social group, which occasionally led
to odd uses of the term. For example, if one shouldn‘t thank slaves, then a free person is
entitled to be thanked.
The modern term ―autonomy‖ doesn‘t really map onto any Greek word, though
―eleutheria‖ ends up doing some of the same work. While the Greeks had a word,
222
H.G. Liddell and R. Scott, Greek-English Lexicon with a Revised Supplement (Oxford: Clarendon Press,
1996), 532. 223
Oxford English Dictionary, "free, adj., n., and adv.", in OED Online
<http://dictionary.oed.com/cgi/entry/50089637> [accessed 1 June 2009].
108
―autonomos‖, the term referred almost exclusively to states and their relationships to each
other. For example, if a state was not ruled by another state, it was ―autonomos‖, literally
self-ruled. The sole classical use of autonomia for a human being is in the Antigone by
Sophocles, in which the chorus criticizes Antigone for behaving like an ―autonomos‖
(641). As this is its sole use to refer to a human being while the political use is very
common, this is likely a metaphor and intended to sound odd. Many of the modern uses
of ―autonomy‖ find themselves under the Greek term ―eleutheria‖, such that if one
wishes to know what Greeks might have thought about autonomy, one should look to
their use of the term ―eleutheria.‖ The modern term, ―autonomy‖ has several definitions,
as identified by Nomy Arpaly:
1. agent autonomy: that I act on my reasons & beliefs.
2. self-efficacy: that I am independent in the world.
3. independence of mind: that I use my own judgement.
4. normative autonomy: that I am a moral agent with moral rights.
5. authenticity: that I am true to myself.224
The first use, agent autonomy, did not usually fall under the heading of ―eleutheria‘, but
under ―enkrateia‖ or ―self-control‖. The cigarette smoker who tries to quit but cannot
lacks agent autonomy, but would not usually be a ―slave‖ or ―doulos‖ in Greek.225
However, ―eleutheria‖ included both the second and third modern uses of ―autonomy‖,
self-efficacy and independence of mind. Self-efficacy was a paradigmatic freeman and
even aristocratic value in ancient Greece. Independence of mind and the ability to make
one‘s own decisions was also very much a part of the values of the freeman, and to a
large extent underlay the traditional synthesis‘s connection between freedom and
persuasion. With respect to the fourth definition, it would be anachronistic to identify the
Kantian connection between agent autonomy and normative autonomy with ―eleutheria‖.
However, there is still a class use of the term that includes rights: the ―eleutheros‖ is
224
This is a short summary of the helpful taxonomy in Nomy Arpaly, "Which Autonomy?," in Freedom
and Determinism (Topics in Contemporary Philosophy) (Cambridge: The MIT Press, 2004), 173-6. 225
André Laks notes an exception to this at Laws 635c-d, but also notes that it is exceptional (André Laks,
"Freedom, Liberty and Liberality in Plato's Laws", Social Philosophy and Policy, 24 [2007]: 150-1). The
line in the Protagoras, translated as ―they think of his knowledge as being utterly dragged around by all
these other things as if it were a slave‖ by Stanley Lombado and Karen Bell actually uses the term
―ἀλδξαπόδνο‖, not ―δνύινο‖, so does not constitute a counterexample (352b).
109
entitled to a number of rights as a member of his or her social class. Finally, the
authenticity definition seems to have no analogous Greek meaning.
I will translate the term ―eleutheria‖ as ―freedom‖ throughout this chapter,
however, it is important to note that the debate about ―eleutheria‖ also bears on modern
debates about autonomy. Specifically, ―eleutheria‖ as a term included both the self-
efficacy and independence of mind definitions of ―autonomy‖. Therefore, discussions of
“eleutheria‖, in so far as they concern the self-efficacy and independence of mind, are
relevant to modern discussions of autonomy. Further, being an ―eleutheros‖ did indeed
bring with it certain rights, though the arguments for this were different. In some ways,
the kind of rights an eleutheros was entitled to can be compared or contrasted to the kinds
of rights autonomous agents are entitled to. Plato‘s conception of ―eleutheria‖, therefore,
is relevant to modern discussions of autonomy, as it does much of the same conceptual
work as ―autonomy‖.
4 The Traditional Synthesis
Traditional Greek thought had many dyads, pairs of opposites set in contrast to
each other. The traditional understanding of the relationship of freedom and persuasion
was the result of a relationship between two of these dyads. The first was freedom
―eleutheria‖ and slavery ―douleia‖. The second was persuasion (―peithō‖) and force
(―bia‖ and sometimes ―anankē‖). These were arrayed in a very specific way. One was
free if and only if one was moved by persuasion, while one was a slave if and only if one
was moved by force. This traditional synthesis was especially important in the
democratic society of Athens whose sense of its own freedom was closely tied to the
persuasion exercised in the democratic assembly. In this section, I will provide several
examples of this traditional synthesis, three from the tragedians and one from an orator.
The first two examples are the Suppliant Maidens by Aeschylus and the Heraclidae by
Euripides. These two plays include a fairly straightforward case in which the freedom of
a city is threatened by force. The third is the more complex use of the terms in the
Philoctetes by Sophocles, in which the class meanings of ―eleutheria‖ are brought to the
110
fore.226
Finally, I will discuss Lysias, a strong democratic partisan, who provides an
argument for the traditional synthesis.
Aeschylus‘s Suppliant Maidens provides an early instance of the traditional
synthesis. In that play, the daughters of Danaus seek asylum with the Argives. The
Argives are presented as proponents of freedom, promoting both speech and persuasion,
while opposing force. There are two passages that are important. The first (605-624) is
that in which the Argives explain why it is they will help the Danaids. The second (941-
949) is when Pelasgus explains to the Egyptian herald why he will not turn over the
maidens to Egpyt. In the first passage, Danaus describes how his daughters received
sanctuary. He says that the Argives, whom he describes as free (―eleutherous‖), held a
vote to allow his daughters to stay after having been persuaded by their king Pelasgus
(609). Here he contrasts the behaviour of the Egyptians, who were using force
(―prostithēi to karteron‖), with that of the Argives, whom Pelasgus persuaded using
speeches (―epeithe rhēsin amph’hēmōn legōn anax Pelasgōn‖) (612,615-616).227
This
passage has several interesting elements. First, note the political anachronism that
Aeschylus has introduced. The Argives were a monarchy. However, because Aeschylus
wished to write a story including themes of freedom and force, he presents the Argives as
having a hybrid government, in which it is run by a king with the consent (and even the
universal votes! (―pantelē psēphismata”)) of the governed (599).228
This anachronistic
use of democracy shows just how deeply the association between democracy and
persuasion was; in order to write about persuasion, Aeschylus needed to introduce an
anachronistic democratic process into the Argive monarchy. Second, notice that the
freedom of the Argives is not simply self-interested. Rather, because the Argives are
free, they defend the freedom of others, that is, they defend the freedom of the maidens.
Freedom is not competitive. In fact, free nations are defenders of others‘ freedom.
Finally, notice just how important speech is to freedom and persuasion. It is the means
by which persuasion and therefore freedom can operate. Danaus emphasizes this when
226
Though Sophocles was born a generation before Euripides, the Philoctetes was written after the
Heraclidae, which is why I discuss them in this order. 227
In this section, I will be citing several, very short passages of Greek. Rather than use cumbersome
footnoting, I will instead include the transliterated text within the body of the section. 228
In The First Democracies, Eric W. Robinson notes the similarities between the political process
described in the Suppliant Maidens and that of fifth-century Athens (Eric W. Robinson, The First
Democracies: Early Popular Government Outside Athens [Stuttgart: Steiner, 1997], 46-47).
111
he uses the redundant phrase, ―talking with speeches‖ (“rhēsin…legōn‖) (615). Speech is
the medium by which freedom and persuasion may operate.
When the Egyptian herald arrives, the traditional synthesis connecting persuasion
to freedom and force to slavery recurs in a speech by King Pelasgus. He allows that the
Egyptians may take the maidens, but only through persuasion (―agois an, eiper eusebēs
pithoi logos‖) (941). He then contrasts the acceptable way by which the herald may take
the maidens with the improper one, trying to take them by force, which will he claims
will never happen, ―mēpot’ekdounai biai” (943). This is an especially clear example of
the persuasion-force dyad, as Pelasgus directly contrasts the two means, the first of which
is acceptable, but the latter of which is not. Having allied himself with persuasion and
against force, King Pelasgus then describes himself as free again, this time in the context
of freedom of speech, saying, ―you hear clearly from a free-spoken tongue‖ (―saphē
d’akoueis ex’eleutherostomou glōssēs‖ (948-9). The Suppliant Maidens provides a clear
instance of the traditional synthesis: the herald may persuade or use force; however, if he
uses force, he will be opposed by the free, democratic and loquacious Argives.
Euripides‘ Heraclidae, written forty years after the Suppliant Maidens, has a quite
similar premise and deals with some of the same themes, though with a different focus.
Rather than focus on the allegiance between persuasion and force, it focuses on the
enmity between freedom and force. In this case, the children of Heracles and their
guardian, Iolaus, seek sanctuary from Eurystheus who desires to kill them. In this case,
they seek sanctuary in Athens, freeing Euripides from the projection of Athenian
government onto other cities (though not from the anachronism). Athens is said to be an
especially free city, and Athens‘ freedom is mentioned several times during the play, at
62 (“eleuthera te gai‖), at 113 (―gēn…eleutheran‖), 198 (―Athēnas tasd’eleutheras‖),
287 (―eleutheran‖), and 957 (―polism’eleutheron‖).229
The first half of the tragedy
provides an example of the political process in Athens and then justifies the Athenian
decision to protect the Heraclidae. Both sides, Iolaus, who is pleading for sanctuary, and
Corpeus, who wants to take the children and threatens to invade Athens if he cannot, are
229
John Wilkins notes that the references in Heraclidae to freedom are all to that of the nation of Athens,
rather than to the freedom of individual members of the society, a use to which Euripides puts the term in
his own Supplices (Euripides, Heraclidae, ed. John Wilkins [Oxford: Clarendon Press, 1979], 57). This
shows that the traditional synthesis applied not only to individuals, but to states as a whole.
112
given an opportunity to plead their respective cases. The chorus mentions how important
it is to hear both stories (―muthoi‖) to render a proper judgement (180). The Athenian
chorus as well as the king, Demophon, are persuaded by the Iolaus, though for different
reasons. The chorus says that they are moved by pity (―Ōiktir’”) for the unfortunate
nobles (232). Demophon himself gives three reasons. The first reason is piety to Zeus, at
whose statue the children are standing, and the second is kinship. However, the third,
which he says is most important (―dei malista phrontisai‖) is the very fact that Corpeus
threatened him (242). He says:
For if I am to allow this altar to be robbed by the force (―biai‖)
of a foreigner, it will be thought that it is no free land
(―eleutheran gaian‖) I govern but that I have betrayed suppliants
for fear of the Argives. (243-5)230,231
Because of the status of Athens as a free country, it cannot surrender the children to
threat of violence. Violence and freedom are inherently incompatible, according to
Demophon, so much so that he cannot give the children to Corpeus precisely because he
was threatened. Though Heraclidae does not discuss persuasion per se, it provides an
example of the kind of persuasion in action. It shows the incompatibility between
freedom and force that was a part of the traditional synthesis and an example of
persuasion in action.
Sophocles‘ Philoctetes uses several examples of the various means of getting
someone to comply, and ranges over force, persuasion and trickery. It discusses the
relationship between force and freedom, though in this case, it considers freedom with
reference to social class and the rights associated with it. Moreover, it considers it not
only in contrast to slavery, but also in constrast to the ―aristos‖ or upper-class nobility.
In the section of the play I will discuss, Neoptolemus and Odysseus have already
succeeded in tricking Philoctetes into giving them the bow of Heracles. They decide to
add insult to injury by compelling Philoctetes to come with them to Troy. The discussion
230
εἰ γὰξ παξήζσ ηόλδε ζπιᾶζζαη βίαη
μέλνπ πξὸο ἀλδξὸο βσκόλ, νὐθ ἐιεπζέξαλ
νἰθεῖλ δνθήζσ γαῖαλ, Ἀξγείσλ δ‘ ὄθλση
ἱθέηαο πξνδνῦλαη· 231
Translations for Euripides are those of Edward P. Coleridge in Euripides, "The Plays of Euripides," in
Great Books of the Western World, ed. Mortimer J. Adler, trans. Edward P. Coleridge (Chicago:
Encyclopædia Brittanica, 1952), 5:203-454.
113
they have has force and its impact as its central theme. Force is mentioned repeatedly
throughout the subsequent conversation, at 983 (―biāi‖), at 988 (―biāi‖), at 990
(―kratōn”) and at 998 (―biāi‖). Philoctetes asks Odysseus and Neoptolemus if they
intend to use force on him (―hoid’ek bias axousin?”) (985). Odysseus gives a somewhat
strange answer, saying that he will use force on him if he does not come willingly (―Ēn
mē herpēis hekōn‖). Odysseus is using ―willing‖ in an odd sense here, because his
statement here implies that a decision made even under duress would still be considered
something that is done ―willingly‖. This is reminiscent of Corpeus‘ speech in the
Heraclidae, who mixed threats and persuasion by threatening the Athenians from within
the assembly. Like Demophon, Philoctetes also does not believe that a decision made
under a threat can be considered free, but instead complains that, ―My father clearly sired
us as slaves (―doulous‖) rather than as freemen (―eleutherous‖)‖ (995-996).232,233
By
referring to his father, Philoctetes is using eleutheros to refer to a social class as well as
simply absence of compulsion, by reminding Odysseus that freedom from compulsion
should be his right by birth. By being subject to force, Philoctetes‘ social status as a
freeman is damaged as it is an entitlement of a freeman not to be threatened.
Odysseus‘ response is a jarring contrast between freemen and aristocrats that
would be of particular resonance to the democracy in Athens. Odysseus retorts that he is
making Philoctetes neither a slave nor a freeman, but an aristos: ―No, but as those similar
to the best (―aristois‖) with whom you must travel to Troy and take it by force (“biāi‖)‖
(997-998). The ―aristoi”, the term the oligarchs normally call themselves, operate by
force, and they intend to take Troy by force (998).234
This would have had special
resonance in the democratic Athens of 409 B.C., written between the oligarchic takeovers
in 411 and 404 B.C. Odysseus here is pointing out a contrast between freedom and force,
but a different one from the standard freedom-slavery contrast. Not only is slavery
connected to force, oligarchy is as well. Oligarchs use force not only against their slaves
but against each other, and Odysseus is requiring the Philoctetes join him in using force
232
... κᾶο κὲλ ὡο δνύινπο ζαθο
παηὴξ ἄξ‘ ἐμέθπζελ νὐδ‘ ἐιεπζέξνπο. 233
This complaint is meant, in a sense, ironically. He is not claiming that he was actually born a slave, but
that Odysseus is treating him as though he was born a slave. In other words, he is complaining that he is
being treated inappropriately for a member of his class. 234
John Carlevale notes this class use of “aristoi‖ in the Philoctetes in Carlevale, "Education, "Phusis," and
Freedom in Sophocles' "Philoctetes"", Arion, Third Series, 8 (2000): 45-6.
114
against yet more ―aristoi‖ in the siege of Troy.235
Philoctetes tries to commit suicide, but
is stopped by Odysseus. He then accuses Odysseus of having no freedom in his mind:
―mēd’eleutheron phronōn‖ (1006). In this passage, Sophocles contrasts freedom both
with slavery and with oligarchy.
I will now turn from the playwrights to one of the most vocal proponents of the
traditional synthesis, the orator, Lysias. Lysias was a passionate partisan of democracy.
Lysias‘s family history makes clear just how strong a democratic partisan he was.
Lysias‘s father, the same Cephalus of the Republic, had been personally invited to live in
the Piraeus by the democratic leader, Pericles. In the oligarchic coup of 404 B.C., he and
his brother, the same Polemarchus of the Republic, were arrested by the oligarchic Thirty,
and Polemarchus was murdered while Lysias managed to escape. Lysias later provided
money, arms and recruitment to the democrat Thrasybulus who successfully seized the
Piraeus and restored democracy to Athens in 403 B.C., granting Lysias a (short-lived)
citizenship in return for his efforts. This support of democracy and hatred of oligarchy
found its way into many of Lysias‘ speeches.236
In the Epitaphius, a showpiece praising those who had fallen in battle serving
Athens, Lysias clearly presents the traditional synthesis. He discusses the origin of
democracy in Athens, associating it, as he often does, with freedom, claiming that the
freedom of all people binds people together and that democracy provides political life
using (or perhaps for) free souls:
They were the first and the only people in that time to drive out
the ruling classes of their state and to establish a democracy,
believing the freedom of all to be the strongest bond of
agreement; by sharing with each other the hopes born of their
perils they had freedom of soul in their civic life. (18) 237
235
This competitive zero-sum game that is aristocratic force will be an important theme in Gorgianic
rhetoric. 236
This short biography borrows its details from S. C. Todd, A Commentary on Lysias: Speeches 1-11
(Oxford: Oxford University Press, 2007). He notes that ―the shadow of 404/3 dominates the corpus‖, but
also notes that Lysias would on occasion represent moderate oligarchs in (unrelated) court cases. 237
πξηνη δὲ θαὶ κόλνη ἐλ ἐθείλῳ ηῶ ρξόλῳ ἐθβαιόληεο ηὰο παξὰ ζθίζηλ αὐηνῖο δπλαζηείαο δεκνθξα-
ηίαλ θαηεζηήζαλην, γνύκελνη ηὴλ πάλησλ ἐιεπζεξίαλ ὁκόλνηαλ εἶλαη κεγίζηελ, θνηλὰο δ‘ ἀιιήινηο ηὰο ἐθ
ηλ θηλδύλσλ ἐιπίδαο πνηήζαληεο ἐιεπζέξαηο ηαῖο ςπραῖο ἐπνιηηεύνλην.
115
In such a system, speech, rather than force, reigns, and this is the appropriate form of
government for human beings. Force, on the other hand, is appropriate only among
beasts:
For they deemed that it was the way of wild beasts to be subject
to one another by force, but the duty of men to delimit justice by
law, to persuade by speech, and to serve these two in act by
submitting to the sovereignty of law and the instruction of
speech. (19)238
In this speech, Lysias provides perhaps the clearest statement of the traditional synthesis,
as well as an argument for why persuasion is better. First, notice that Lysias is
connecting freedom with democracy in much the same way as the Aeschylus and
Euripides did by using democracies as the settings for their discussions of force and
persuasion. Next, he says that democracy is connected with ―free souls‖. The speech has
an ambiguous dative that could mean that democracy rules ―with‖ free souls (a dative of
means) or ―for‖ free souls (a dative of interest). Either way, democracy is allied with
freedom. Next, Lysias clearly separates force and persuasion, clearly allying persuasion
with democracy and freedom. This persuasion works through ―logos‖, which I have
translated as ―speech‖. Though ―reason‖ is also a possible translation of the term, I have
chosen ―speech‖ as it is the term consistently used to refer to speech in the context of
oratory.239
By connecting persuasion to speech, he echoes the connection is Aeschylus
between between free speech and persuasion.
In addition, Lysias gives not only an account of the traditional synthesis, but an
argument as well. Subjection by force, he says, is not appropriate to human beings
because subjection by force is the way that beasts operate. Instead, he says, because we
are not beasts, we ought instead to persuade each other through speech and write our
decisions in laws. In fact, his position is actually stronger than the traditional synthesis.
Force is not just appropriate to slaves. Force is appropriate to beasts.240
Human beings
238
γεζάκελνη ζεξίσλ κὲλ ἔξγνλ εἶλαη ὑπ‘ ἀιιήισλ βίᾳ θξαηεῖζζαη, ἀλζξώπνηο δὲ πξνζήθεηλ λόκῳ κὲλ
ὁξίζαη ηὸ δίθαηνλ, ιόγῳ δὲ πεῖζαη, ἔξγῳ δὲ ηνύηνηο ὑπεξεηεῖλ, ὑπὸ λόκνπ κὲλ βαζηιεπνκέλνπο, ὑπὸ ιόγνπ
δὲ δηδαζθνκέλνπο. 239
For example, the historical Gorgias in Encomium of Helen, the author of the Hippocratic On the Art and
Aristophanes in The Clouds all use the term in this way. 240
Isocrates will later make this same association, arguing that, ―because there has been implanted in us the
power to persuade each other and to make clear to each other whatever we desire, not only have we
escaped the life of wild beasts, but we have come together and founded cities and made laws and invented
116
using force on one another is quite literally dehumanizing and it resembles the strategy of
animals.241
Rather, because we have speech, something that separates us from the beasts,
we ought to use that special power for persuasion. Animals, which have no speech, can
never persuade. They can only threaten and be violent. We can then record the decisions
produced by those discussions in the form of law, something else that the beasts cannot
do. So, Lysias grounds his defense of democracy on the uniquely human power of
speech. Moreover, he connects three of the most important parts of the traditional
synthesis, freedom and persuasion and their opposition to force. In his version of the
synthesis, beastliness takes the role of slavery, which is in many ways an even stronger
claim.
Variations on the traditional synthesis were repeated throughout the tragedians
and in the rhetor Lysias. In those works, freedom and persuasion were closely connected,
while force was put in opposition to them. Lack of freedom was characterized in
different ways, including slavery, shame and beastliness. In addition, some of the
political context is evident in these works. Freedom and persuasion were seen to be
closely allied with democracy, the usual system of government in classical Athens, and
Sophocles explicitly contrasted it with oligarchy. Moreover, some other observations
about the traditional synthesis were made by these authors. Aeschylus saw freedom as
inherently self-diffusing, and a free state will not only defend its own freedom, but that of
others. Both Euripides and Sophocles noted that even threats of violence were a danger
to freedom. Lysias noted the connections not only with speech but also with law.
Despite some small differences, these works had a great deal in common. Not only was
arts‖ (―κῖλ ηνῦ πείζεηλ ἀιιήινπο θαὶ δεινῦλ πξὸο κᾶο αὐηνὺο πεξὶ ὧλ ἂλ βνπιεζκελ, νὐ κόλνλ ηνῦ
ζεξησδο δῆλ ἀπειιάγεκελ, ἀιιὰ θαὶ ζπλειζόληεο πόιεηο ᾠθίζακελ θαὶ λόκνπο ἐζέκεζα θαὶ ηέρλαο
εὕξνκελ, θαὶ ζρεδὸλ ἅπαληα ηὰ δη᾽ κλ κεκεραλεκέλα ιόγνο κῖλ ἐζηηλ ὁ ζπγθαηαζθεπάζαο‖).
Translation taken from Isocrates, "Nicocles or the Cyprians," in Isocrates With an English Translation in
Three Volumes, ed. George Nolin (London: William Heinemann Ltd., 1980), 1:74-115. 241
Hesiod also connects law and humanity in Works and Days when he says ―For the son of Cronos has
ordained this law for men, that fishes and beasts and winged fowls should devour one another, for right is
not in them; but to mankind he gave right which proves far the best‖:
―ηόλδε γὰξ ἀλζξώπνηζη λόκνλ δηέηαμε Κξνλίσλ,
ἰρζύζη κὲλ θαὶ ζεξζὶ θαὶ νἰσλνῖο πεηεελνῖο
ἔζζεηλ ἀιιήινπο, ἐπεὶ νὐ δίθε ἐζηὶ κεη‘ αὐηνῖο·
ἀλζξώπνηζη δ‘ ἔδσθε δίθελ, ἡ πνιιὸλ ἀξίζηε
γίλεηαη.‖
Note, however, that he does not include speech or persuasion in his equation. Translation is from Hesiod,
"Works and Days," in The Homeric Hymns and Homerica with an English Translation, ed. Hugh G.
Evelyn-Wight (London: Harvard University Press, 1914), 45-65.
117
there no possible conflict between persuasion and freedom, the two were inextricably
connected. Further, persuasion and force were diametrically opposed. It was these
elements of the traditional synthesis that Gorgias challenges, and I now turn to Gorgias to
show the challenge that this synthesis faced.
5 Gorgias’ Challenge
5.1 The Historical Gorgias
The boasts of rhetors like Gorgias provided a serious threat to the traditional
synthesis. Gorgias outright boasted that he was able to teach rhetoric so well that it
became a kind of force that could enslave others. If Gorgias was right in this claim and
persuasion was really a sort of force, then the traditional synthesis breaks down entirely.
If persuasion is really force, then the freedom of those who are persuaded is but an
illusion. Instead, it is merely slavery in disguise. This would imply that no one who is
moved by another in any way is free and that true freedom, if it exists at all, is the
competitive freedom that comes from controlling others. The traditional synthesis, then,
would be a false synthesis. Not only was what Gorgias was teaching a threat to the
traditional synthesis, but to democracy. On the one hand, it threatened democratic
Athens‘ sense of its own freedom by arguing that there is no difference between being
persuaded and being compelled. On the other hand, Gorgias, by teaching his skill for
money, put persuasion in the hands of the wealthy elite, giving them increasing power in
the city. If the Gorgianic claim was true and he taught rhetoric so well that it became a
kind of force, it would spell the end of the traditional synthesis.
I will focus on Gorgias‘s speech Encomium of Helen, which provides not only
Gorgias‘s strongest claims about the power of rhetoric but also provides some insight into
why he believed that rhetoric was so powerful. In this showpiece, Gorgias defends
Helen, who had abandoned her husband, Menelaus, to join Paris, sparking the Trojan
War.242
Gorgias provides four possible explanations for her departure: she was
compelled by Fate, she was kidnapped by Paris, she was persuaded by speech or she was
242
By a ―showpiece‖, I mean a speech that was not intended to be used for its stated purpose, such as a
defense of the long-deceased Helen.
118
in love. In all of these cases, he claims that Helen would have been coerced; hence, she
is not responsible. Gorgias claims that these four possibilities are exhaustive, and
therefore, Helen ought not to be blamed for going to Troy.243
The case of kidnapping is
obvious, and the cases of fate and love are cases of divine compulsion by the Fates and
Aphrodite, respectively. However, the inclusion of persuasion is particularly important
here. If Helen went to Troy as a result of persuasion in speech, Gorgias claims that she
was just as coerced as if Paris had kidnapped her. I will quote this passage in full as it
includes several important differences with the traditional synthesis, and I will refer back
to it over the course of this and the next section:
What cause then prevents the conclusion that Helen similarly,
against her will, might have come under the influence of speech,
just as if a forceful person seized her by force. For it was
possible to see how the force of persuasion prevails; persuasion
has the form of necessity, but it does not have the same power.
For speech constrained the soul, persuading it which it
persuaded, both to believe the things said and to approve the
things done. The persuader, like a constrainer, does the wrong
and the persuaded, like the constrained, in speech is wrongly
charged.244,245
This passage unravels the traditional synthesis thread by thread. First, notice his claim
that Helen might have been persuaded against her will. This is, in many ways, an odd
claim, as one would think that it would be impossible to be persuaded of something
against one‘s will; it is precisely one‘s will that is changed. However, according to
Gorgias, someone persuaded cannot be said to be acting willingly, which would imply
that they cannot be said to be acting freely. Gorgias then explicitly denies the traditional
contrast between persuasion and force. Persuasion, he says, can act as a ―forceful
person‖ (“biatērion‖) seizing by force (―bia‖). He then describes the force of persuasion
243
Scott Cosigny argues that Gorgias here is using a ―tree division‖, in which he uses a division to
aggressively rule out alternatives, something that he notes that, done properly, is a part of true dialectic and
rhetoric in Phaedrus 265d-e (Scott Consigny, Gorgias: Sophist and Artist [Columbia: University of South
Carolina Press, 2001], 187-88). 244
ηίο νὖλ αἰηία θσιύεη θαὶ ηὴλ ιέλελ ὕκλνο ἤιζελ ὁκνίσο ἂλ νὐ λέαλ νὖζαλ ὥζπεξ εἰ βηαηήξηνλ
βία ξπάζζε. ηὸ γὰξ ηῆο πεηζνῦο ἐμῆλ ὁ δὲ λνῦο θαίηνη εἰ ἀλάγθε ὁ εἰδὼο ἕμεη κὲλ νὖλ, ηὴλ δὲ δύλακηλ ηὴλ
αὐηὴλ ἔρεη. ιόγνο γὰξ ςπρὴλ ὁ πείζαο, ἡλ ἔπεηζελ, λάγθαζε θαὶ πηζέζζαη ηνῖο ιεγνκέλνηο θαὶ ζπλαηλέζαη
ηνῖο πνηνπκέλνηο. ὁ κὲλ νὖλ πείζαο ὡο ἀλαγθάζαο ἀδηθεῖ, δὲ πεηζζεῖζα ὡο ἀλαγθαζζεῖζα ηη ιόγση κάηελ
ἀθνύεη θαθο. 245
Unless otherwise noted, translations of Gorgias are that of George Kennedy from Rosamond Kent
Sprague, ed., The Older Sophists (Indianapolis: Hackett, 1973).
119
(―to...tēs peithōus‖) and says that it wears the mask of compulsion (―anankē‖). The
wordplay in this passage makes it evident that Gorgias knows how paradoxically he is
speaking. Rather than accept the traditional contrast between persuasion and force,
Gorgias argues that persuasion is a sort of force.
However, he says, persuasion does not have the same power (“dynamis‖) as
necessity. In this context, the word seems to mean something like the means by which
persuasion compels. Rather than use the traditional tools of compulsion, threats and
bodily manipulation, persuasion does two things: it causes people to believe the things
said and to approve of the things done. Specifically, then, persuasion influences what we
believe and what we praise. Through this influence of what we believe to be true and
what we believe to be best, the rhetor can force us to do whatever it is he or she wants us
to do. This was the case with Helen, who was persuaded to approve of running off with
Paris and then did so. So, while persuasion does not actually grab us or threaten us, it can
and does entice us to believe and act in certain ways.
If Gorgias is right, then the traditional synthesis is as mistaken as a synthesis
could possibly be. Not only would persuasion not be allied with freedom against force,
but persuasion would be allied with force against freedom. The traditional synthesis
would not only be false; it would be a sham. Moreover, democratic Athens, in which
people are proud that they are moved by persuasion rather than force, would remain
unfree.
5.2 Plato’s Gorgias
Plato, too, presents Gorgias as threatening this traditional synthesis, though not
quite as starkly as the historical Gorgias. Plato wrote a dialogue named the Gorgias, the
first part of which features Gorgias in a discussion with Socrates. In that dialogue,
Gorgias appears as being at least apparently a supporter of the traditional synthesis.
However, Socrates quickly uncovers that Gorgias is an opponent of the traditional
synthesis. The character Gorgias first argues that rhetoric is the source of freedom for
human beings, echoing what sounds like a Lysian sentiment.246
However, it quickly
246
As there is an important distinction between the historical Gorgias and the eponymous character that
appears in the dialogue, I herein shall refer to the former as ―the historical Gorgias‖ and the latter as ―the
120
becomes clear that he believes that rhetoric is a competitive craft, one with winners and
losers. Under the traditional synthesis, the distinction between freedom and slavery is a
passive one. If I am persuaded, I am free, but if I am forced, I am a slave. On the other
hand, the freedom that the character Gorgias discusses is an active one. Rhetoric
provides freedom by teaching others to persuade. It is a competitive craft and contributes
to the competitive freedom and self-efficacy of the aristocrat. Therefore, for every free
persuader, there is an enslaved persuaded and freedom loses its self-effusiveness. The
character Gorgias, despite his apparent agreement with the traditional synthesis, therefore
turns out to be as much a threat to the traditional synthesis as his historical counterpart.
At first, the character Gorgias makes a claim that would have been right at home
in one of Lysias‘s speeches. He says that rhetoric is ―…the source of freedom for
humankind itself...‖ (452d).247
So far so good. That persuasion would be the source of
freedom for human beings fits well with the traditional synthesis. However, immediately
trouble starts to arise. The sentence continues, ―...and at the same time it is for each
person the source of rule over others in one‘s city‖ (452d).248
This is not incompatible
with the Lysian position. After all, rulers over free people are those who have
successfully persuaded them. However, notice that the promised benefit is no longer for
―human kind itself‖ ( ―autois tois anthropois‖), but for a smaller number of rulers, which
he calls ―for each‖ (―hekastōi‖). Still, the traditional synthesis allows for leaders and
despite the worry that the wealthy will be the only ones able to afford to learn rhetoric
and be leaders, so long as people are persuaded rather than forced, they would still be
considered free. However, the character Gorgias quickly reveals that he does not hold the
Lysian view but one much closer to the historical Gorgias, when he makes the following
claim about rhetoric: ―In point of fact, with this ability you‘ll have the doctor for your
slave, and the physical trainer, too‖ (452e).249
At this point, the character Gorgias reveals
that he has nothing like the traditional synthesis in mind. When someone is persuaded
character Gorgias‖. So that this might not become cumbersome, I will do this only once per paragraph so
long as I do not refer to both in that paragraph. 247
αἴηηνλ ἅκα κὲλ ἐιεπζεξίαο αὐηνῖο ηνῖο ἀλζξώπνηο 248
ἅκα δὲ ηνῦ ἄιισλ ἄξρεηλ ἐλ ηῇ αὑηνῦ πόιεη ἑθάζηῳ. 249
θαίηνη ἐλ ηαύηῃ ηῇ δπλάκεη δνῦινλ κὲλ ἕμεηο ηὸλ ἰαηξόλ, δνῦινλ δὲ ηὸλ παηδνηξίβελ·
121
using rhetoric, that person is enslaved according to Gorgias.250
Rather than being a
source of freedom for the persuaded, rhetoric is a source of enslavement
At this point, it is a little unclear how to even interpret the character Gorgias‘s
earlier boast that rhetoric was the source of freedom ―for humankind itself‖. Either
Gorgias is simply contradicting himself, and Plato‘s intention is to show that Gorgias and
other rhetors are deeply confused about their crafts, or when Gorgias said that rhetoric
was the source of freedom for humankind itself, he did not mean that it made everyone
free.251
Instead, Gorgias perceives rhetoric as a competitive craft, one which has winners
and losers and only the winners turn out to have rule and be free.252
Later in the dialogue,
Gorgias argues that rhetoric, like any competitive skill (―tēi agōniāi”), can be used either
for good or evil. The introduction of competition here is very important. Competitions
have winners and losers, and not everyone is able to successfully persuade. Those who
are able to persuade are those who will have rulership (―archein‖) within the city. Those
who are unable to persuade will not have rulership. The worst off will be those
persuaded, who will be enslaved by the winner of the rhetorical competition. Gorgias
conceives of rhetoric as a form of competition with winners and losers, both among the
rhetors and among their hearers. The prize for winning this competition is rulership.
This does not accord with the traditional synthesis, because the traditional synthesis
maintained the freedom not only of the successful persuader, but also of the persuaded.
As a result, by treating freedom as self-efficacy and rhetoric as a means of attaining that
self-efficacy, Gorgias denies an important part of the traditional synthesis: persuasion
gives freedom to the persuaded as well as the persuader.
250
Plato will return to this claim of Gorgias in the Philebus, where Protarchus relates Gorgias‘s claim that
―the art of persuasion is superior to all others because it enslaves all the rest, with their own consent, not by
force, and is therefore by far the best of all the arts‖ (― ηνῦ πείζεηλ πνιὺ δηαθέξνη παζλ ηερλλ—πάληα
γὰξ ὑθ‘ αὑηῇ δνῦια δη‘ ἑθόλησλ ἀιι‘ νὐ δηὰ βίαο πνηνῖην, θαὶ καθξῶ ἀξίζηε παζλ εἴε ηλ ηερλλ‖) (58a-
b). In this passage, the character Gorgias is contrasting what is done willingly (―δη‘ ἑθόλησλ‖) with what is
done by force (―δηὰ βίαο‖), unlike in the historical Gorgias in the Encomium of Helen, where he says that
persuasion is a type of force (―ὥζπεξ εἰ βηαηήξηνλ βία ξπάζζε‖). Nonetheless, the language of slavery
(―δνῦια‖) from the Gorgias recurs here. 251
This would be in the same way as one might say that Christianity is the source of salvation for
humankind without necessarily implying that everyone is saved. 252
Rachel Barney discusses the agonistic nature of Gorgianic rhetoric and Gorgias‘ defense of it in Rachel
Barney, "Gorgias' Defense, Plato and His Opponents on Rhetoric and the Good", The Southern Journal of
Philosophy, 48 (2010): 95-121.
122
Both the historical Gorgias and the character Gorgias provide serious challenges
to the traditional synthesis. The historical Gorgias challenged the traditional antithesis
between force and persuasion, equating the two and denying the possibility of human
freedom in the presence of persuasion. The character Gorgias altered the definition of
freedom, treating it as a form of competitive self-efficacy, of which persuasion was a tool
and thereby asserting that someone persuaded was a slave. Together, both Gorgiases
denied nearly every part of the traditional synthesis, and I will refer to these denials as
―the Gorgianic challenge‖.
6 Plato’s Response
6.1 Persuasion and Magic
Despite his usual hostility to the historical Gorgias and the sophists, Plato partly
accepts the Gorgianic challenge to the traditional synthesis. Specifically, he accepts
Gorgias‘s claim that there is a type of persuasion under which one cannot be said to be
free. Plato even adopts Gorgias‘ sorcery analogy, which I will discuss in this section and
uses it in several dialogues. This ―sorcerous‖ persuasion works by mixing pleasure or
fear with belief so as to manipulate those beliefs. Therefore, the traditional synthesis is
incorrect. To be persuaded is not to be free, if the persuasion involved is sorcerous
persuasion. Where Plato dissents from Gorgias, however, is on the nature of freedom.
Gorgias treats persuasion as a zero-sum game, in which the persuader defeats both the
persuaded and other persuaders. Plato instead argues that there is a criterion by which
one persuaded can be considered free, knowledge. When one has knowledge, one is
immune to sorcerous persuasion. The competitive craft of persuasion is replaced with the
diffusive craft of teaching as the source of freedom and knowledge becomes its new
criterion.
I will begin this section by discussing the method of Gorgianic persuasion and the
sorcery analogy he used to describe it. I will then show how Plato adopts this analogy.
The term the historical Gorgias and Plato use is ―goēteia”, which can mean ―sorcery‖,
123
―trickery‖ or even ―jugglery‖.253
I will use ―sorcery‖ to translate this term throughout, as
its primary meaning is sorcery, while the use for ―trickery‖ is a metaphor. The Greek-
English Lexicon cites ―sorcerer, wizard‖ as the primary meaning of goēs, the related
noun, and ―witchcraft, jugglery‖ as the first two definitions of goēteia (356). As such,
the primary referent of the term is magical powers, while the use of the term to mean
merely ―trickery‖ is secondary. Plato, uses the term as one referring to magical powers.
In the Symposium, Diotima lists the powers given by daemons to human beings, and puts
goēteia on the list with sacrifices, mysteries, enchantments and prophesy (202e-203a).254
Given the supernatural context of the term, it is important not to kill the metaphor
prematurely by translating the term ―trickery‖ even in those cases where Plato is clearly
using the term metaphorically. To do so would be to miss the force of Plato‘s metaphor,
and perhaps to imply a more mundane meaning when Plato may have even meant its use
literally. Rather than beg any interpretive questions of when Plato intends his use of the
term metaphorically, I will translate it literally as ―sorcery‖ for the noun and as
―ensorcel‖ for the verb.
In order to understand the sorcery metaphor that persists throughout both the
historical Gorgias and Plato, one must understand the method of Gorgianic persuasion.
In the Encomium of Helen, Gorgias argues that speech is ―a powerful lord‖ (―dunastēs
megas‖) (8) that can manipulate emotions and by manipulating those emotions, cause
someone to change his or her beliefs. As evidence for this he provides the example of
poetry. Poetry, which he defines as speech with meter, can cause ―fearful shuddering and
tearful pity and grievous longing‖ (―phrikē periphobos kai eleos poludakrus kai pothos
philopenthēs‖) because, when people encounter the suffering of others, they tend to share
that suffering (9). Moreover, the ability to change someone‘s emotional state gives one
the ability to change his or her opinions. Being able to create pleasure allows one to
merge those emotions with the opinions of the hearers:
Sacred incantations sung with words are bearers of pleasure and
banishers of pain, for, merging with opinion in the soul, the
253
―Jugglery‖ itself is something of an archaic English term, referring to ―The art or practice of a juggler;
minstrelsy, play; pretended magic or witchcraft; conjuring, legerdemain‖ (Oxford English Dictionary,
"jugglery", in OED Online <http://dictionary.oed.com/cgi/entry/50124579> [accessed 2 September 2010]). 254
δηὰ ηνύηνπ θαὶ καληηθὴ πᾶζα ρσξεῖ θαὶ ηλ ἱεξέσλ ηέρλε ηλ ηε πεξὶ ηὰο ζπζίαο θαὶ ηειεηὰο θαὶ ηὰο
ἐπῳδὰο θαὶ ηὴλ καληείαλ πᾶζαλ θαὶ γνεηείαλ.
124
power of the incantation is wont to beguile it and persuade it and
alter it by sorcery (―goēteiai‖). (10)255
It becomes clear as this section of the Encomium continues that the sacred incantations
here are not those of wizards but of anyone who uses speech in order to merge pleasure
and pain with opinions. All of the examples of this sorcery that Gorgias provides are
examples of Gorgianic persuasion, including Paris‘s seduction of Helen. Gorgias further
compares persuasion to a kind of ―drug‖ or ―pharmakon‖ (14). Just as a drug produces
health and disease, so can a speech produce emotions:
For just as different drugs dispel different secretions from the
body, and some bring an end to disease and others to life, so also
in the case of speeches, some distress, other delight, some cause
fear, others make hearers bold, and some drug and ensorcel
(―exegoēteusan‖) the soul with a kind of evil persuasion. (14)256
Though ―drug‖ or ―epharmakeusan‖ here is a medical term, and the analogy is
specifically medical, this may not be a mixed metaphor. ―Pharmakon‖ has ―enchanted
potion, philter; hence charm, spell‖ as its third definition in the Greek-English Lexicon
(1917), so the phrase ―drug and ensorcel‖ is not as odd as it appears. Gorgias is a little
vague as to the exact mechanism by which one moves from emotional manipulation to
change in opinion, but the key seems to be in his phrase ―merging with opinion in the
soul‖ (10). It is evident that, in moral matters, the thought of something with pleasure is
more likely to make us approve of the thing done. However, Gorgias also believes that
rhetoric can work in non-moral matters, and he includes the examples of astronomy and
philosophy as those easily influenced by persuasion (13). In the Encomium of Helen,
then, Gorgias develops an extended magic analogy throughout the work, claiming that
persuasion is a kind of sorcery that manipulates belief through manipulating the emotions
surrounding that belief.
Plato borrows the historical Gorgias‘s use of ―sorcery‖ in reference to persuasion
and often accuses those who practice sophistry and rhetoric of being ―goētai”,
specifically, illusionists and enchanters. Plato may even have the last line of the above
255
αἱ γὰξ ἔλζενη δηὰ ιόγσλ ἐπσηδαὶ ἐπαγσγνὶ δνλῆο, ἀπαγσγνὶ ιύπεο γίλνληαη· ζπγγηλνκέλε γὰξ ηῆη
δόμεη ηῆο ςπρῆο δύλακηο ηῆο ἐπσηδῆο ἔζειμε θαὶ ἔπεηζε θαὶ κεηέζηεζελ αὐηὴλ γνεηείαη. 256
ὥζπεξ γὰξ ηλ θαξκάθσλ ἄιινπο ἄιια ρπκνὺο ἐθ ηνῦ ζώκαηνο ἐμάγεη, θαὶ ηὰ κὲλ λόζνπ ηὰ δὲ βίνπ
παύεη, νὕησ θαὶ ηλ ιόγσλ νἱ κὲλ ἐιύπεζαλ, νἱ δὲ ἔηεξςαλ, νἱ δὲ ἐθόβεζαλ, νἱ δὲ εἰο ζάξζνο θαηέζηεζαλ
ηνὺο ἀθνύνληαο, νἱ δὲ πεηζνῖ ηηλη θαθῆη ηὴλ ςπρὴλ ἐθαξκάθεπζαλ θαὶ ἐμεγνήηεπζαλ.
125
passage in mind, ―and some drug and ensorcel‖, when he has Diotima list the qualities of
the daemon, Eros, claiming that he is ―a clever sorcerer (―goēs”), potion-maker
(―pharmakos‖) and sophist (―sophistēs) (Symposium 203d)‖. One way that Plato argues
that sophists manipulate the emotions of hearers is through imitation. As illusionists,
sophists and rhetors are said to be ―goētai‖ by creating illusions or imitations that
ensorcel those who view the imitation. Plato characterizes sophists as illusionists in three
different dialogues, the Sophist, the Statesman and the Euthydemus. In the Sophist, the
Eleatic Visitor claims that the Sophist ought to be called a goēs and also an imitator
(―mimētēs‖) who ensorcels by putting words in the ears:
Well then, won‘t we expect that there‘s another kind of expertise
– this time having to do with words – and that someone can use
it to ensorcel young people when they stand even further away
from the truth about things? (234c)257
Moreover, in the Statesman, the one who puts forward a false constitution, not grounded
in expertise, is ―the greatest sorcerer among the sophists‖ (291c3). In the Euthydemus,
Socrates claims that Euthydemus and Dionysodorus are imitating (―mimeisthon‖) the
sorcery (goēteuonte) of the Egyptian sophist (―sophistēn‖) Proteus, a deity who refused to
assume his true shape until wrestled by Menelaus (288b, cf. Odyssey iv:456ff.). Plato
thereby borrows Gorgias‘ use of ―sorcery‖ and attributes it to the sophists.
Moreover, Plato describes as sorcery a kind of persuasion that directly
manipulates the emotions of the hearer through pleasure without imitation.258
The
clearest example of this is in the Menexenus, in which Socrates describes the effect that a
speech he just heard had on his soul:
They do their praising so splendidly that they ensorcel our souls
(goēteuousin), attributing to each individual man, with the most
257
{ΞΔ.} Σί δὲ δή; πεξὶ ηνὺο ιόγνπο ἆξ‘ νὐ πξνζδνθκελ εἶλαί ηηλα ἄιιελ ηέρλελ, ᾗ αὖ δπλαηὸλ <ὂλ>
[αὖ] ηπγράλεη ηνὺο λένπο θαὶ ἔηη πόξξσ ηλ πξαγκάησλ ηῆο ἀιεζείαο ἀθεζηηαο δηὰ ηλ ὤησλ ηνῖο ιόγνηο
γνεηεύεηλ. 258
An open question is whether or not Plato intended to say that all sorcerous persuasion was necessarily
also mimetic. Christopher Janaway, for example, conflates the two, saying that ―Plato apparently agrees
with Gorgias that speech, when used by mimetic poetry, drugs and bewitches the soul‖ (Christopher
Janaway, Images of Excellence: Plato's Critique of the Arts [Oxford: Oxford University Press, 1999], 143).
However, the mimetic language is Plato‘s, not Gorgias‘s. On the one hand, the three-fold division of
involuntary loss of beliefs in Republic III implies that there may be a more direct, non-mimetic way of
producing beliefs through pleasure that is distinct from other sorts of deception. Hence, young people must
be exposed to fears and pleasures, ―testing them more thoroughly than gold is tested by fire‖ (413d). On
the other hand, Gorgias 464-465 includes the claim that rhetoric mimics politics, implying that all
production of belief through pleasure is mimetic.
126
varied and beautiful verbal embellishments, both praise he merits
and praise he does not, extolling the city in every way, and
praising the war-dead, all our ancestors before us, and us
ourselves, the living. The result is, Menexenus, that I am put
into an exalted frame of mind when I am praised by them. Each
time, as I listen and fall under their spell, I become a different
man – I‘m convinced that I have become taller and nobler and
better looking all of a sudden. (235a-b)259
Note what is happening here. The rhetor is using beautiful words that praise people both
deservingly and undeservingly. Undeserved praise is obviously false. However, because
the words are so beautiful, Socrates cannot help but believe what is being said. His
opinions, then, are being changed simply by virtue of the beauty of the words, in some
cases to opinions that are false. Specifically, in this case it is the belief that the object of
praise is, in fact, deserving of that praise. This is a quite similar process to that described
in the Encomium of Helen. By producing pleasure in the soul surrounding a particular
belief, a rhetor can make someone likely to hold that belief. In fact, Socrates not only
begins to believe undeserved praise, he starts to hold different opinions about himself.
He believes that he is nobler and better looking, both of which are difficult to measure,
but he even begins to believe he is taller, something measurable and obviously false.
Praise then changes the opinions of the hearers by causing pleasure in the hearer, and
Plato, like Gorgias, calls this a sort of magic.
Plato claims in the Phaedo that this kind of change of opinion through pleasure is
a result of our body and its effect on our soul. In that case, the belief that the only things
that exist are physical is a result of the ensorcelling of the soul by our bodies when we
experience too much pleasure and desire:
But I think that if the soul is polluted and impure when it leaves
the body, having always been associated with it and served it,
bewitched (goēteuomenē) by physical desires and pleasure to the
259
γνεηεύνπζηλ κλ ηὰο ςπράο, θαὶ ηὴλ πόιηλ ἐγθσκηάδνληεο θαηὰ πάληαο ηξόπνπο θαὶ ηνὺο
ηεηειεπηεθόηαο ἐλ ηῶ πνιέκῳ θαὶ ηνὺο πξνγόλνπο κλ ἅπαληαο ηνὺο ἔκπξνζζελ θαὶ αὐηνὺο κᾶο ηνὺο ἔηη
δληαο ἐπαηλνῦληεο, ὥζη‘ ἔγσγε, ὦ Μελέμελε, γελλαίσο πάλπ δηαηίζεκαη ἐπαηλνύκελνο ὑπ‘ αὐηλ, θαὶ
ἑθάζηνηε ἐμέζηεθα ἀθξνώκελνο θαὶ θεινύκελνο, γνύκελνο ἐλ ηῶ παξαρξῆκα κείδσλ θαὶ γελλαηόηεξνο
θαὶ θαιιίσλ γεγνλέλαη.
127
point at which nothing seems to exist for it but the physical…
(81b)260
Plato here provides another example of an opinion changed through pleasure. The
pleasures and desires we encounter in the body trick us into believing that the physical
world is the only real world, rather than the world of ideas. One should note that even if
one does not accept the existence of an intellectual world, this is independent from the
psychological mechanism Plato is observing. Plato is noting that pleasures and desires
create beliefs in us. In the Menexenus case, pleasure created approval of whatever is
praised. In the Phaedo case, pleasure creates belief in the ontological primacy of the
objects of pleasure and desire. Pleasure and desire, then, alter opinion by altering what
we believe to be good and even what we believe to be ultimately real.
6.2 Knowledge and Freedom
However, Plato still believes that freedom is possible, and the key to that freedom
is knowledge. The historical Gorgias had largely equated what he taught with persuasion
as a whole, but Plato argues that there are two types of persuasion: the first is the
sorcerous type of persuasion, where beliefs are generated through the manipulation of the
emotions; the second, however, is teaching, in which one provides knowledge to one‘s
subject. In the Gorgias, Socrates makes this distinction to the character Gorgias, who
then admits that there is such a distinction, and that Gorgias does not engage in the type
of persuasion involving teaching:
SOCRATES: Would you like to posit two types of persuasion, one
providing conviction without knowledge, the other providing
knowledge?‖
GORGIAS: Yes, I would.
SOCRATES: Now which type of persuasion does oratory produce
in law courts and other gatherings concerning things that are just
and unjust? The one that results in being convinced without
knowledge or the one that results in knowing?
260
ὰλ δέ γε νἶκαη κεκηαζκέλε θαὶ ἀθάζαξηνο ηνῦ ζώκαηνο ἀπαιιάηηεηαη, ἅηε ηῶ ζώκαηη ἀεὶ ζπλνῦζα
θαὶ ηνῦην ζεξαπεύνπζα θαὶ ἐξζα θαὶ γνεηεπνκέλε ὑπ‘ αὐηνῦ ὑπό ηε ηλ ἐπηζπκηλ θαὶ δνλλ, ὥζηε
κεδὲλ ἄιιν δνθεῖλ εἶλαη ἀιεζὲο ἀιι‘ ἠ ηὸ ζσκαηνεηδέο.
128
Gorgias: It‘s obvious, surely, that it‘s the one that results in
conviction. (454e)261
Socrates then takes his argument another step. What he notes is that the kind of
persuasion in which Gorgias is engaged cannot work on those who already know the truth
about a subject. After Gorgias boasts that he is capable of persuading concerning health
when doctors are not persuasive and even that he could be appointed state physician over
a doctor, Socrates makes the following observation:
SOCRATES: And doesn‘t ‗in a gathering‘ just mean ‗among those
who don‘t have knowledge‘? For, among those who do have it, I
don‘t suppose that he‘ll be more persuasive than the doctor.
GORGIAS: That‘s true. (459a)262
Plato has not simply put words into the mouth of Gorgias here, as the historical Gorgias
makes this exact distinction in the Encomium. There, he tries to explain why it is that his
type of persuasion is able to work and makes the claim that it is because people have only
beliefs, and not knowledge:
For if all men on all subjects had both memory of things past and
awareness of things present and foreknowledge of the future,
speech would not be similarly similar, since as things are now it
is not easy for them to recall the past nor consider the present nor
predict the future. So that on most subjects most men take
opinion as counselor to their soul. (11)263
What Gorgias is arguing here is that, if people had knowledge of all subjects in all times,
speech would have no effect on them. However, because people lack knowledge, they
form beliefs instead. Beliefs, though, are the ground in which sorcerous persuasion can
work, by mixing in pleasure and fear. Therefore, the historical Gorgias himself admits
that those who know are immune to his form of persuasion.
261
{Ω.} Βνύιεη νὖλ δύν εἴδε ζκελ πεηζνῦο, ηὸ κὲλ πίζηηλ παξερόκελνλ ἄλεπ ηνῦ εἰδέλαη, ηὸ δ‘
ἐπηζηήκελ; {—ΓΟΡ.} Πάλπ γε. {—Ω.} Πνηέξαλ νὖλ ῥεηνξηθὴ πεηζὼ πνηεῖ ἐλ δηθαζηεξίνηο ηε θαὶ ηνῖο
ἄιινηο ὄρινηο πεξὶ ηλ δηθαίσλ ηε θαὶ ἀδίθσλ; ἐμ ἥο πηζηεύεηλ γίγλεηαη ἄλεπ ηνῦ εἰδέλαη ἠ ἐμ ἥο ηὸ εἰδέλαη;
{—ΓΟΡ.} Γῆινλ δήπνπ, ὦ ώθξαηεο, ὅηη ἐμ ἥο ηὸ πηζηεύεηλ. 262
{—Ω.} Οὐθνῦλ ηὸ ἐλ ὄριῳ ηνῦηό ἐζηηλ, ἐλ ηνῖο κὴ εἰδόζηλ; νὐ γὰξ δήπνπ
ἔλ γε ηνῖο εἰδόζη ηνῦ ἰαηξνῦ πηζαλώηεξνο ἔζηαη.
{—ΓΟΡ.} Ἀιεζῆ ιέγεηο. 263
εἰ κὲλ γὰξ πάληεο πεξὶ πάλησλ εἶρνλ ηλ <ηε> παξνηρνκέλσλ κλήκελ ηλ ηε παξόλησλ <ἔλλνηαλ> ηλ
ηε κειιόλησλ πξόλνηαλ, νὐθ ἂλ ὁκνίσο ὅκνηνο ἤλ ὁ ιόγνο, νἷο ηὰ λῦλ γε νὔηε κλεζζῆλαη ηὸ παξνηρόκελνλ
νὔηε ζθέςαζζαη ηὸ παξὸλ νὔηε καληεύζαζζαη ηὸ κέιινλ εὐπόξσο ἔρεη· ὥζηε πεξὶ ηλ πιείζησλ νἱ πιεῖζηνη
ηὴλ δόμαλ ζύκβνπινλ ηῆη ςπρῆη παξέρνληαη.
129
Through Socrates‘ argument, Plato is challenging the historical Gorgias in two
ways. First, Socrates is dividing up persuasion in a way that Gorgias did not. Gorgias
himself lumped all persuasion together under his own sorcerous persuasion. Socrates is
pointing out that there is a second form of persuasion, teaching. Unlike persuasion,
teaching is not a zero-sum, competitive craft, as one does not get the better of the person
taught when they are taught: the person taught benefits. As a result, Gorgias is incorrect
to conclude that all persuasion is of the sort he teaches. The mistake that both the
traditional synthesis and Gorgias make is to treat all persuasion as equivalent. The
traditional synthesis treats all persuasion as being freeing, while Gorgias treats it all as
being coercive. Instead, Plato divides persuasion itself into categories, only one of which
is the kind of sorcerous persuasion that Gorgias teaches. The new category is not
―trickery‖ at all, but teaching.
Second, teaching and knowledge provide a perfect antidote to Gorgias‘s sorcerous
persuasion. As both the character and historical Gorgias admitted, having knowledge
makes one immune to the kind of persuasion that he uses. As Principle #2: Crafts and
knowledge are co-extensive implies that someone with a craft has knowledge, having a
craft makes one immune to Gorgias‘s persuasion. He cannot trick doctors concerning
medicine, only patients, and only patients who do not understand medicine at that. His
kind of sorcerous persuasion can only mix itself with belief (―doxa‖), not knowledge
(―epistēmē‖). What this implies is that those who have knowledge are free in an
important way, that is, in the sense that they have independence of mind. No amount of
emotional manipulation can cause someone to lose one‘s knowledge. Plato thereby
shows that Gorgias‘s understanding of persuasion is both incomplete and vulnerable to an
antidote.
Plato‘s most explicit connection of knowledge and freedom is in the Lysis. In that
dialogue, Socrates speaks to his very young friend, Lysis, who complains that his parents
will not allow him to ―do what he likes‖ (―poiein hoti an boulēi‖) (208e). Socrates
notices that Lysis‘s parents do not let him drive the chariots or the mules and that those
jobs are entrusted instead to slaves instead, while Lysis is hindered from performing
those tasks. Socrates points out there is a discrepancy between the supposed class
structure and what people are allowed to do in practice. The slaves, though they are
130
slaves, are permitted to engage in a number of tasks, while Lysis, who is ―well-born‖
(―gennaiou‖ 209a) is permitted to do very few things. Lysis provides a very class-based
response, and he tries to point out that what is occurring is in accord with the class
structure. The apparent discrepancy is actually in accord with current social norms
because Lysis has not yet come of age (―Ou gar pō, ephē, hēlikian echō‖ 209a).
However, Socrates then argues that it is not Lysis‘ lack of age that has caused him
to be hindered from performing the tasks he likes. He points out that there are a number
of things, including writing or playing the lyre that he is allowed to do, despite his not yet
having come of age. Socrates then asks Lysis why it is that he is allowed to perform
writing and flute playing, while he is prevented from driving the mules, and Lysis
provides the desired answer: ―I suppose it‘s because I understand these things and not
those‖ (209c).264
This is not simply a convention-based response such as Lysis had
previously given. Instead, it is providing the natural criterion according to which tasks
are divided: knowledge. Socrates agrees with Lysis, and points out that this is true even
in the case of the crown prince of Persia. Even the crown prince of Persia would not be
allowed to tinker with a stew unless he knew how to cook (209d-e).
Knowledge, Socrates concludes, provides the freedom to do what we want. He
then makes a claim highly reminiscent of Gorgias‘ claim that rhetoric is the source of
freedom in the Gorgias. Socrates says the following:
Then this is the way it is, my dear Lysis: in those areas where we
really understand something everybody – Greeks and barbarians,
men and women – will trust us, and there we will act just as we
want, and nobody will want to get in our way. There we will be
free (―eleutheroi‖) ourselves, and in control of others (―allōn
archontes‖). (210a-b) 265
Here Socrates makes an even stronger claim about the relationship between knowledge
and freedom than he made in the Gorgias. In that dialogue, knowledge makes one
immune to the potential manipulation of sorcerous oratory. However, here, having
knowledge provides one with the trust of others that enables us to do what we want. Like
264
Ὅηη νἶκαη, ἔθε, ηαῦηα κὲλ ἐπίζηακαη, ἐθεῖλα δ‘ νὔ. 265
Οὕησο ἄξα ἔρεη, ἤλ δ‘ ἐγώ, ὦ θίιε Λύζη· εἰο κὲλ ηαῦηα, ἃ ἂλ θξόληκνη γελώκεζα, ἅπαληεο κῖλ
ἐπηηξέςνπζηλ, Ἕιιελέο ηε θαὶ βάξβαξνη θαὶ ἄλδξεο θαὶ γπλαῖθεο, πνηήζνκέλ ηε ἐλ ηνύηνηο ὅηη ἂλ
βνπιώκεζα, θαὶ νὐδεὶο κᾶο ἑθὼλ εἶλαη ἐκπνδηεῖ, ἀιι‘ αὐηνί ηε ἐιεύζεξνη ἐζόκεζα ἐλ αὐηνῖο θαὶ
ἄιισλ ἄξρνληεο,
131
the slave cook who is allowed to spice the sauce instead of the king‘s son, when people
think we have knowledge, we are free to act with regard to the things we know.
The passage connects with the dialogue Gorgias in two very important and
striking ways. First, the language that Socrates uses here is very similar to the language
that the character Gorgias used in the Gorgias. In the Lysis, Socrates claims that we are
free (―eleutheroi‖) and in control of others (―allōn archontes‖) when we have knowledge.
Notice, though, that Gorgias makes a similar boast using the same terms in the Gorgias.
There, he says that rhetoric is ―…the source of freedom (―eleutherias‖) for humankind
and at the same time it is for each person the source of rule over others (―allōn archein‖)
in one‘s own city‖ (452d). The parallel between these passages is striking, and may be an
example of Plato altering a sophist‘s slogan in order to give it a contrary or different
meaning.266
The character Gorgias and probably the historical Gorgias claim that
rhetoric is the source of freedom and rule. Plato uses this same claim, but replaces
rhetoric with knowledge. When one takes someone else‘s claim and switches a single
term, it provides heavy emphasis to that term, effectively saying, ―Instead of rhetoric,
knowledge is the source of freedom and rule.‖267
This claim here includes an implicit
rejection of Gorgias‘s position.268
Plato also ties philosophy and the class of freemen together in the Theaetetus and
the Sophist, where he addresses the class-based question of the sort of education that is
proper to a freeman. In the Protagoras, Socrates asks his ecstatic young friend
Hippocrates whether Protagoras taught what was appropriate to freemen: ―you didn‘t get
from them [Hippocrates‘ teachers] a technical instruction to become a professional, but a
general education suitable to a freeman (―ton eleutheron‖)‖ (312b).269,270
Hippocrates
266
Perhaps the most famous instance of this is the comment of the Athenian stranger that ―God is the
measure of all things‖ at Laws 716c, a clear revision of Protagoras‘s ―man is the measure of all things‖
(Rosamond Kent Sprague, ed., The Older Sophists [Indianapolis: Hackett, 1973], Fr. 1). This appears to be
a similar reworking of someone else‘s slogan. 267
Note also that the combination of freedom and rule points toward Priniciple #6: Crafts are arranged
in a hierarchy, in which tool-supplying crafts are subordinate to those that use the tools. Having
knowledge puts one in a position of proper rulership over anyone with a subordinate craft. 268
If the slogan that the character Gorgias uses in the Gorgias is actually a slogan of the historical Gorgias,
then nothing can be inferred about the relative chronology of the Gorgias and the Lysis. However, if the
slogan in the Gorgias is Plato‘s own invention and the Lysis is actually quoting the Gorgias, it would imply
that the Lysis was written after the Gorgias. 269
ηνύησλ γὰξ ζὺ ἑθάζηελ νὐθ ἐπὶ ηέρλῃ ἔκαζεο, ὡο δεκηνπξγὸο ἐζόκελνο, ἀιι‘ ἐπὶ παηδείᾳ, ὡο ηὸλ
ἰδηώηελ θαὶ ηὸλ ἐιεύζεξνλ πξέπεη.
132
responds in the affirmative. The education of a freeman is not craft-based, according to
Hippocrates, but is based on understanding things that are free from technical application.
A similar question about what is appropriate to teach a freeman appears in the Laches,
this time in military education: ―this [fighting in armor] and horsemanship are forms of
exercise especially suited to a freeman‖ (182a).271
In this dialogue, Laches and Nicias are
debating what kind of military instruction is appropriate for their sons, who are freemen.
Finally, in the (admittedly spurious) Rival Lovers, Socrates and a young, unnamed friend
debate over the value of polymathy, learning a little bit about everything, versus a more
narrow philosophical training. The young friend, a defender of polymathy, defends it
explicitly on the grounds that it is suitable to a freeman:
The most admirable and proper sorts of learning are those from
which one derives the most fame as a philosopher, and one
acquires the most fame by appearing to be an expert in all the
skills, or if not in all of them, in most of the really important
ones, learning as much of them as is proper for a free man.
(135b)272
According to Socrates‘ young friend, then, a philosopher should learn about all of the
skills, within the constraints of what is appropriate for a freeman to learn. As one can
see, the question of what is appropriate teaching for the freeman was a live topic
throughout the dialogues.
Surprisingly, in the Theaetetus and the Sophist, Socrates and the Eleatic visitor
make the claim that the appropriate subject of learning for a freeman is philosophy,
specifically dialectic!273
Knowledge, specifically philosophical knowledge, thereby
becomes the criterion of the freeman class. These dialogues are a part of a trilogy of
dialogues recounting a single discussion, starting with the Theaetetus, then continuing
with the Sophist and concluding with the Statesman.274
In the Theaetetus, Socrates
compares the life of a philosopher to that of someone in the assembly, the place that
someone like the character Gorgias trains people to speak. Paradoxically, Socrates
270
This is not the physician Hippocrates, but another man with the same name. 271
θαὶ ἅκα πξνζήθεη κάιηζη‘ ἐιεπζέξῳ ηνῦηό ηε ηὸ γπκλάζηνλ θαὶ ἱππηθή. 272
ἂλ πιείζηελ δόμαλ ἔρνη ηηο εἰο θηινζνθίαλ· πιείζηελ δ‘ ἂλ ἔρνη δόμαλ, εἰ δνθνίε ηλ ηερλλ ἔκπεηξνο
εἶλαη παζλ, εἰ δὲ κή, ὡο πιείζησλ γε θαὶ κάιηζηα ηλ ἀμηνιόγσλ, καζὼλ αὐηλ ηαῦηα ἃ πξνζήθεη ηνῖο
ἐιεπζέξνηο καζεῖλ, 273
It may not be so surprising, however, in light of Principle #4: Crafts are dependant on dialectic for a
determinate subject matter and to systematically approach causal relationships. 274
An implied fourth part, named The Philosopher, was never written.
133
argues that rhetors like Gorgias are actually enslaved to those they are trying to persuade,
while the philosopher is free from concern about the passions of others. In fact, Socrates
says, rhetors like Gorgias are like slaves, while philosophers are like freemen:
SOCRATES: Well, look at the man who has been knocking about
in law courts and such places ever since he was a boy; and
compare him with the man brought up in philosophy, in the life
of a student. It is surely like comparing the upbringing of a slave
with that of a freeman. (172c-d) 275
Socrates seems to be directly challenging the claims of rhetors like Gorgias. Gorgias is
not teaching people to be free, but rather to be slaves. Instead, philosophy is the true
source of freedom.
Those skilled at speaking in the law courts, Socrates claims, are in a constant
position of needing to flatter the desires of the audience, in order to get what they want
from them. Constantly appealing to the passions of others in courts is slavish, according
to Socrates. This slavery takes three different forms. First, the person in the law courts is
a slave to the procedures of the law courts. While the philosopher is able to speak on any
subject in search of what is, the person in the law courts is required to be a slave to the
current topics and the clock: ―But the other – the man of the law courts – is always in a
hurry when he is talking; he has to speak with one eye on the clock‖ (172d-e).276
Second,
law courts operate under compulsion (“anankē‖), in which one person sues another.
Someone defending themselves in a lawcourts is compelled to respond to a charge point-
by-point:
…he has his adversary standing over him, armed with
compulsory powers and with the sworn statement, which is read
out point by point as he proceeds, and must be kept to by the
speaker. The talk is always about a fellow-slave, and is
addressed to a master, who sits there holding some suit or other
in his hand. (172e)277
The life of the court room, therefore, is apparently a slavish life.
275
{Ω.} Κηλδπλεύνπζηλ νἱ ἐλ δηθαζηεξίνηο θαὶ ηνῖο ηνηνύηνηο ἐθ λέσλ θπιηλδνύκελνη πξὸο ηνὺο ἐλ
θηινζνθίᾳ θαὶ ηῇ ηνηᾷδε δηαηξηβῇ ηεζξακκέλνπο ὡο νἰθέηαη πξὸο ἐιεπζέξνπο ηεζξάθζαη. 276
νἱ δὲ ἐλ ἀζρνιίᾳ ηε ἀεὶ ιέγνπζη—θαηεπείγεη γὰξ ὕδσξ ῥένλ. 277
θαὶ νὐθ ἐγρσξεῖ πεξὶ νὗ ἂλ ἐπηζπκήζσζη ηνὺο ιόγνπο πνηεῖζζαη, ἀιι‘ ἀλάγθελ ἔρσλ ὁ ἀληίδηθνο
ἐθέζηεθελ θαὶ ὑπνγξαθὴλ παξαλαγηγλσζθνκέλελ ὧλ ἐθηὸο νὐ ῥεηένλ [ἡλ ἀλησκνζίαλ θαινῦζηλ]· νἱ δὲ
ιόγνη ἀεὶ πεξὶ ὁκνδνύινπ πξὸο δεζπόηελ θαζήκελνλ, ἐλ ρεηξί ηηλα δίθελ ἔρνληα.
134
So far, it would appear that Socrates is speaking exclusively about the court
system, specifically of Athens. That a defendant is required to respond to a charge point-
by-point is apparently a part of Athenian legal procedure. Gorgias claims not only to be
able to teach people to speak in courtrooms, but also in political assemblies or even in
private. Further, Socrates seems to be envisioning only being sued, not bringing a suit.
The compulsion that a defendant experiences is the compulsion that a prosecutor wields.
However, when he describes what it is that the lawyer is skilled at, he quickly reveals that
is concern is broader than just defenders in the Athenian courtroom. First, he notes that
the skill involved in being good at public speaking is that of flattery, something which is
actually a skill honed very well by slaves in powerless positions: ―Such conditions make
him keen and highly strung, skilled in flattering the master and working his way into
favor; but cause his soul to be small and warped‖ (173a).278
Moreover, this requirement
that we learn how to flatter our masters is a part of any public speaking, including the
assemblies and even including playwrights. Theodorus, with whom Socrates is speaking,
quickly draws this general conclusion: ―We [philosophers] have no jury and no audience
(as the dramatic poets have), sitting in control over us, ready to criticize and give orders‖
(173c).279
Because someone speaking in an assembly is appealing to other people, they
are acting in a slavish way, flattering their audience and trying to please them.
What Socrates is engaged in here is a clever bit of class commentary. He is using
some very important observations about slavery in order to accuse the rhetors and
sophists of behaving slavishly. As many slaves, abused and exploited people have
discovered, the ability to flatter those who have power over them is an important part of
making slavery tolerable. One learns to please one‘s master so as not to be beaten. By
accusing sophists and rhetors of teaching flattery, Socrates is accusing them of teaching
them skills only needed by slaves. A nobleman and a freeman don‘t need to flatter. They
are independent, and don‘t need to be obsequious. Rather, flattery implies that the person
being flattered has the real power. By learning to flatter the crowds in order to get them
to do what he wants, Gorgias is actually teaching his students to treat crowds as slaves
treat their masters. This doesn‘t teach freedom; it teaches slavery. Socrates uses the
278
ὥζη‘ ἐμ ἁπάλησλ ηνύησλ ἔληνλνη θαὶ δξηκεῖο γίγλνληαη, ἐπηζηάκελνη ηὸλ δεζπόηελ ιόγῳ ηε ζσπεῦζαη θαὶ
ἔξγῳ ὑπειζεῖλ, ζκηθξνὶ δὲ θαὶ νὐθ ὀξζνὶ ηὰο ςπράο. 279
νὔηε γὰξ δηθαζηὴο νὔηε ζεαηὴο ὥζπεξ πνηεηαῖο ἐπηηηκήζσλ ηε θαὶ ἄξμσλ ἐπηζηαηεῖ παξ‘ κῖλ.
135
traditional class distinctions in Athens to show that the skill that Gorgias is claiming to
teach is actually not an appropriate skill for a free person at all.
This claim that philosophy is the education appropriate to a freeman continues in
the Sophist, in this case specifically focusing on the dialectical method that is developed
in that dialogue. In their search for the sophist, the visitor and Theatetus happen on a
method of collecting and division that they decide isn‘t appropriate to a sophist, and the
Eleatic visitor makes the following claim:
Or for heaven‘s sake, without noticing have we stumbled on the
knowledge that freemen have? Maybe we‘ve found the
philosopher even though we were looking for the sophist.
(253c)280
This is a very important claim. First, we see something of an answer to the lingering
question of what kind of education is appropriate to a freeman. The appropriate
knowledge for a freeman is dialectic. Moreover, the Eleatic visitor actually equates the
freeman and the philosopher in this passage. The person who has dialectic is both a true
freeman and a true philosopher. The knowledge of dialectic, which is the knowledge that
a true philosopher has, is the knowledge of freemen. What a freeman should be learning,
according to this passage, is dialectic, the same knowledge that makes someone a
philosopher. Again, we see a strong connection between freedom and knowledge in
Plato‘s dialogues, this time as a replacement in the curriculum for the lessons of the
sophists.
In several dialogues Plato sets out a new criterion for freedom: knowledge.
Knowledge provides freedom because it makes one immune to the very manipulation that
someone like Gorgias claims to teach. Further, craft knowledge will give us the ability to
act as we want in situations we know something about. Finally, a philosopher‘s
education, in which knowledge is gained through dialectic, is more appropriate to the free
person than one centred on public approval, as the latter turns the speaker into a kind of
slave, pandering to the emotions of his or her audience. Plato thereby sets up knowledge
as a new criterion for freedom.
280
ἠ πξὸο Γηὸο ἐιάζνκελ εἰο ηὴλ ηλ ἐιεπζέξσλ ἐκπεζόληεο ἐπηζηήκελ, θαὶ θηλδπλεύνκελ δεηνῦληεο ηὸλ
ζνθηζηὴλ πξόηεξνλ ἀλεπξεθέλαη ηὸλ θηιόζνθνλ;
136
7 Medical Applications
In this section, I will consider the specifically medical applications of Plato‘s
connections between knowledge and freedom. A number of the examples of the
relationship between knowledge and freedom in the Platonic corpus are medical
examples. In the Gorgias, there is an example of sorcerous persuasion being used in a
medical context. In the Laws, there is an example of the type of understanding that is
appropriate to a free patient. Comparing these passages to some passages in the
Hippocratic corpus, I will attempt to uncover the specfic medical applications and even
the type of knowledge that Plato believes is appropriate for a free patient to know as
opposed to a slave patient.
7.1 Sorcerous Persuasion in Medicine
The case of sorcerous persuasion appears in the Gorgias. In that dialogue, the
character Gorgias wishes to provide an example of the power of rhetoric by giving the
example of his brother, who is a doctor. Gorgias claims that he has gone many times
with either his brother or other doctors and has been able to persuade their patients to
undergo some painful procedure, using solely oratory and with no knowledge of
medicine at all:
Many a time I‘ve gone with my brother or with other doctors to
call on some sick person who refuses to take his medicine or
allow the doctor to perform surgery or cauterization on him.
And when the doctor failed to persuade him, I succeeded, by
means of no other craft than oratory. (456b)281
This passage is interesting for a number reasons. First, Gorgias is not a doctor. In fact,
part of this boast is that he does not need to know medicine in order to persuade people in
crafts of which one has no knowledge. What this implies is that the kind of persuasion
that Gorgias uses in this context cannot be the didactic persuasion that communicates
knowledge. Gorgias lacks knowledge, so that is impossible. Rather, this is purely
sorcerous persuasion. Second, Gorgias claims that he has done this many a time
281
πνιιάθηο γὰξ ἢδε ἔγσγε κεηὰ ηνῦ ἀδειθνῦ θαὶ κεηὰ ηλ ἄιισλ ἰαηξλ εἰζειζὼλ παξά ηηλα ηλ
θακλόλησλ νὐρὶ ἐζέινληα ἠ θάξκαθνλ πηεῖλ ἠ ηεκεῖλ ἠ θαῦζαη παξαζρεῖλ ηῶ ἰαηξῶ, νὐ δπλακέλνπ ηνῦ
ἰαηξνῦ πεῖζαη, ἐγὼ ἔπεηζα, νὐθ ἄιιῃ ηέρλῃ ἠ ηῇ ῥεηνξηθῇ.
137
(―pollakis‖). Gorgias isn‘t providing simply an anecdote. This is something that he has
done often.
Most importantly, the doctors themselves invite the character Gorgias along with
them to persuade a recalcitrant patient, after they themselves have failed to persuade
them. In this passage, Gorgias is not simply boasting about the power of sorcerous
rhetoric to persuade a recalcitrant patient, he is denying the power of didactic persuasion
to persuade a recalcitrant patient. The doctor is only able to transmit information about
health and how to improve the health of the patient. It is unclear in the story why the
patients are recalcitrant, though it is likely for one of two reasons. Either the patient does
not believe that the treatment will really work because the patient‘s fear of pain is
undermining the belief he or she would otherwise have that the treatment will be
beneficial, or the patient does believe the treatment will work, but doesn‘t want to
undergo the treatment because of the pain. In both of these cases, the medical craft has
proven useless in persuading the patient. Fear of pain has mixed with opinion in such a
way that either the patient cannot believe that the treatment will work or the patient
doesn‘t care whether the treatment will work. Rhetoric, on the other hand, is a craft of
mixing pleasure and fear with opinion, according to the historical Gorgias. The character
Gorgias intervenes and manipulates the opinions of the patient using sorcerous
persuasion, unmixing the fear from opinions surrounding the procedure. In other words,
the doctor cannot persuade the patient, because the doctor only understands health. Only
a rhetor can persuade the patient, because only the rhetor understands pleasure and fear.
In a recalcitrant patient, didactic persuasion simply isn‘t relevant because knowledge isn‘t
the problem. Fear is the problem.
Despite the character Gorgias‘s boasts that he can assist physicians, the references
to oratory that do exist in the Hippocratic corpus are largely hostile, precisely because of
the way in which oratory and showmanship could take advantage of patient‘s relative
ignorance about health.282
Due to the lack of any sort of medical accreditation in the
Fifth Century, physicians were largely forced to convince others of their credentials in an
282
In this section I do not intend to imply any more unity to the Hippocratic corpus as a part of a single
school of medicine. Instead, I use them to provide examples of the way in which rhetoric was impacting on
medical practice in ancient Greece.
138
open market.283
However, the methods of medicine were especially unsuited to
promoting their craft in an open setting. Doctors were finding that sorcerous persuasion
was undermining their ability to gain patients. Although sorcerous persuasion might be
used to persuade a recalcitrant patient, the overall opinion of the authors of the
Hippocratic corpus was that sorcerous persuasion usually promoted quackery, at the
expense of good medicine.
There are a number of instances of frustration with other doctors and the methods
by which they gained patients. The author of Law was very concerned about the number
of charlatans in Greece and also with the way in which doctors are chosen. Much of this
work is a polemic against other doctors, but the author is ostensibly concerned not only
with uncovering the imputed deception of other physicians, but also with the ignorance of
patients with respect to the medical craft and the effect that has. Lack of knowledge of
the medical craft leads many patients to be prone to sorcerous persuasion, just as both
Plato and the historical Gorgias had claimed. Coupled with the number of people
pretending to the medical craft, this creates a situation in those claiming to be serious
doctors had difficulty attracting patients and the reputation of the medical profession as a
whole was harmed.
The author of Law believes that medicine has been disgraced (―adoxiēs‖) because
of the ignorance both of those who attempt the art of medicine and of those who choose
doctors (1).284
There are many who claim to be doctors, but very few who are in fact
(1).285
He is claiming that both a) there are a large number of false doctors, and b) the
ultimate problem is with the patients who are unable to distinguish false doctors from
true ones. Moreover, false doctors often attract patients by using showy methods,
designed to impress the patients using spectacle to delight them, rather than any sort of
medical craft. Two such examples of this showy persuasion appear in the work On
Joints. The author of On Joints criticises those who make complex, beautiful nose
bandages without knowing what they are doing (―tēisi kalēisin epidesesin aneu noou‖)
283
G.E.R. Lloyd argues that the lack of accreditation made ancient physicians especially concerned about
their reputations, both because it was difficult to establish oneself as a physician, and also because one‘s
reputation was especially fragile (Hippocrates, Hippocratic Writings, ed. G. E. R. Lloyd [Middlesex:
Penguin Books Ltd., 1978], 14-15). 284
δηὰ δὲ ἀκαζίελ ηλ ηε ρξενκέλσλ αὐηῇ, θαὶ ηλ εἰθῆ ηνὺο ηνηνύζδε θξηλόλησλ. 285
νὕησ θαὶ ἰεηξνὶ, θήκῃ κὲλ πνιινὶ, ἔξγῳ δὲ πάγρπ βαηνί.
139
(35).286
Apparently, it was common for physicians to use elaborate bandages for broken
noses and perhaps other breaks, as the elaborateness of the bandaging would seem to
advertise the skill of the physician. Another example is not only showy and public, but
also outright dangerous. He criticizes those who tie hump-backed patients to a ladder and
drop them suddenly to the ground in hopes of straightening the spine:287
For, to begin with, successions on a ladder never straightened
any case, so far as I know, and the practitioners who use this
method are chiefly those who want to make the vulgar herd gape,
for to such it seems marvelous to see a man suspended or shaken
or treated in such ways. (42)288
These doctors were a threat because, in the face of patients unable to understand
medicine themselves, they turned medicine into showmanship. Both of these cases from
On Joints have in common that they are done publicly. While bandages may not be
applied publicly, they are worn publicly. The gable-dropping apparently was actually
performed publicly, causing the crowd to marvel at what they were seeing. In both of
these cases, the author is criticizing is the susceptibility of a crowd without knowledge to
being impressed with wonders: sorcerous persuasion. At places in the Hippocratic
corpus, the doctors criticize persuasion directly. On the Art is largely a defense against
those who have the ―art of bad speech‖ (―logōn ou kalōn technēi”) and who attempted to
bring shame on physicians who do not rely on rhetoric (On the Art 1). By using the
phrase ―art of bad speech‖, the author is specifically connecting false doctors with that of
deceptive rhetors and sophists.289
Further, in On the Art, the author makes the claim that
what genuine doctors do is to provide a demonstration or ―epideixis‖ of deeds, not
286
Jacques Jouanna raises the possibility that the origin of the spectacle of bandages lay in the historic
connection of bandages with the warrior class, such as Achilles‘ binding of Patroclus‘ wound at Iliad
11.832. Bandaging had historically been something in which warriors themselves were trained and was in
a way glorious (Jacques Jouanna, Hippocrates [Baltimore: Johns Hopkins University Press, 1999], 94-
95,440). However, it is equally plausible that bandages took on this showy role because they were a visible
and enduring sign of the physician‘s craft. 287
Jouanna arges that Hippocratic doctors believed that, in principle, ladders could be used to correct a
deformed spine in a treatment called ―succession‖. However, the dropping of the patient and the publicity
appear to be what is being criticized here (Jacques Jouanna, Hippocrates [Baltimore: Johns Hopkins
University Press, 1999], 97). 288
Σνῦην κὲλ γὰξ, αἱ ἐλ ηῇ θιίκαθη θαηαζείζηεο νὐδέλα πσ ἐμίζπλαλ, ὧλ γε ἐγὼ νἶδα· ρξένληαη δὲ νἱ ἰεηξνὶ
κάιηζηα αὐηῇ νὗηνη νἱ ἐπηζπκένληεο ἐθραπλνῦλ ηὸλ πνιὺλ ὄρινλ· 289
A very similar phrase is used with reference to Socrates (presented as a sophist) in Aristophanes‘ Clouds
in which Phaedippides asks to be taught the ―worse speech‖ (ηὸλ ἣηηνλα) (114). The Worse Speech then
appears as a character in this play.
140
words.290
The term ―epideixis‖ is a technical term from rhetoric, referring to a rhetorical
performance.
The problem for many doctors is that their method of persuasion did not work
very well in a public context. Instead, many doctors rely on building up a reputation
through making correct predictions.291
Their primary method of building up a reputation,
perhaps oddly, was not necessarily curing people but by forecasting what would happen.
This is most thoroughly discussed in the treatise Prognostics. The author characterizes
himself as a prophet, declaring the past, present and future of the patient and his or her
illness:
For if he discover and declare unaided by the side of his patients
the present, the past and the future, and fill in the gaps in the
account given by the sick, he will be the more believed to
understand the cases, so that men will confidently entrust
themselves to him for treatment. (1)292,293
One key element in prognostication was to say whether or not a patient would die. First
of all, one appears wiser when one is able to predict the future.294
More importantly,
however, the doctor will not be blamed for a death if the doctor predicts the death in
advance (1).295
In fact, the author of On the Art goes so far as to argue that doctors
should not try to treat those who will be overmastered by their disease (3)296
.
While this method of prediction worked privately, it was a method particularly
unsuited to the public marketplace. Predictions take time to be fulfilled, and the doctor‘s
reputation would need to be spread largely by word of mouth. Making a proper
290
νἵ ηε λῦλ ιεγόκελνη ιόγνη δεινῦζηλ αἵ ηε ηλ εἰδόησλ ηὴλ ηέρλελ ἐπηδείμηεο, ἃο ἐθ ηλ ἔξγσλ ἣδηνλ ἠ ἐθ
ηλ ιόγσλ ἐπηδεηθλύνπζηλ, 291
An important exception to this is Polybius‘s On the Nature of Man, which describes itself as a lecture.
However, this public interest in medical topics seems to mainly be restricted to ―the ultimate constituents of
the human body‖ rather than the whole of the medical art (Hippocrates, Hippocratic Writings, ed. G. E. R.
Lloyd [Middlesex: Penguin Books Ltd., 1978], 38). 292
These numbers refer to the paragraph in which the quoted text is found. 293
πξνγηγλώζθσλ γὰξ θαὶ πξνιέγσλ παξὰ ηνῖζη λνζένπζη ηά ηε παξεόληα θαὶ ηὰ πξνγεγνλόηα θαὶ ηὰ
κέιινληα ἔζεζζαη, ὁθόζα ηε παξαιείπνπζηλ νἱ ἀζζελένληεο ἐθδηεγεύκελνο, πηζηεύνηη‘ ἂλ κᾶιινλ
γηγλώζθεηλ ηὰ ηλ λνζεόλησλ πξήγκαηα, ὥζηε ηνικᾷλ ἐπηηξέπεηλ ηνὺο ἀλζξώπνπο ζθέαο ἑσπηνὺο ηῶ
ἰεηξῶ. 294
This characterization of knowing past and future is the same language used by Gorgias when he
describes the knowledge that would make one immune to sorcerous persuasion: ―For if men on all subjects
had <both> memory of things past and <awareness> of things present and foreknowledge of the future,
speech would not be similarly similar‖ (Encomium of Helen 11). 295
ηνὺο ἀπνζαλνπκέλνπο ηε θαὶ ζσζεζνκέλνπο πξνγηγλώζθσλ θαὶ πξναγνξεύσλ ἀλαίηηνο ἂλ εἴε. 296
ηὸ κὴ ἐγρεηξέεηλ ηνῖζη θεθξαηεκέλνηζηλ ὑπὸ ηλ λνζεκάησλ.
141
prediction is particular to each patient‘s circumstance and can often take a long time. The
author of Prorrhetics (Predictions) II discusses this process of making predictions.
Making proper predictions requires learning all of the details297
(2), and one must be
careful (2)298
. One needs to take into account the diet and exercise of the patient,
especially if the patient is bed-ridden (3)299
. Fevers can result from changes in diet (3)300
.
As a result, predictions are particular to the course of each patient and disease (3)301
.
This provides a fundamental problem for doctors that rely on prediction in the context of
the Greek marketplace. They do not wish to be dependent on effects or speeches in
words, but their means of gaining reputation, correct prognosis, is always limited to
particular patients observed usually in their homes over long periods of time.
That the public did not understand the medical art made it difficult for physicians
who rely on observation and prediction to attract patients. Instead, those able to impress
patients with public displays undermined them. The observational method of doing
medicine was especially unsuited to public displays, as doctors built up reputations over
time through prognostication rather than presentation. These physicians are concerned
that public displays are undermining their ability to apply their craft.
7.2 Didactic Persuasion in Medicine
In contrast to this use of sorcerous persuasion in the Gorgias, Plato argues instead
that what is appropriate for a free patient is that he or she be educated about his or her
illness, in so far as is possible. Specifically, it is appropriate for a free patient to be
taught about his or her illness in much the same way that the physician himself or herself
understands the illness. The patient should learn both the principles of the illness and
how this illness has come to be. The information should be case-specific, and not
presented simply as a series of rules or ―doctor‘s orders‖ to follow. This passage is tied
together with other passages from the Phaedrus and the Laws, and I will discuss those
relationships as well.
297
ρξὴ πξνιέγεηλ θαηακαλζάλνληα πάληα ηαῦηα. 298
Ὧλ δὴ ἕλεθα θειεύσ ζσθξόλσο ηὰ πξνῤῥήκαηα πνηέεζζαη. 299
θαί ηνη θαηάθεηληαί γε νἱ ἄλζξσπνη θαὶ δηαηηήκαζηλ ὀιηγνηξόθνηζη ρξληαη. 300
ηῶ πνηῶ θαὶ ηνῖζη πεξὶ ηὰ ζηηία ἀκέηξσο ρξήζαηην. 301
Ἀιι‘ ὅκσο πξόζζελ ἠλ κὴ ηὰ ἢζεα ηλ λνζεκάησλ ηε θαὶ ηλ ἀιγεόλησλ ἐθκάζῃ ὁ ἰεηξὸο, νὐ ρξὴ
πξνιέγεηλ νὐδέλ.
142
That didactic persuasion is appropriate to a free patient is presented most clearly
in Book IV of the Laws. In that book, the Athenian visitor provides a long comparison of
two types of medicine: free medicine and slave medicine. The visitor makes the
apparently uncontroversial claim that there are two types of medicine, free and slave
medicine. Free medicine is performed by doctors, and is usually performed on free
patients. However, there is also another sort of medicine that Socrates calls slave
medicine. The doctor‘s slave assistants, who have picked up some medical techniques
through observation without understanding them, will use those techniques on other
slaves, who presumably cannot afford to pay for the doctor themselves. The Athenian
visitor describes these slave doctors in the following way:
Now here‘s another thing you notice. A state‘s invalids include
not only free men but slaves too, who are almost always treated
by other slaves who either rush about on flying visits or wait to
be consulted in their surgeries…Then he dashes off on his way
to the next slave-patient, and so takes off his master‘s shoulders
some of the work of attending the sick. (720c-d)302,303
The slaves, unable to afford the attendance of the physician and not valuable enough to
have a physician paid for by their masters, are treated instead by other slaves, in this case,
the slaves of the doctors. However, these slave doctors are not doctors per se.
Specifically, they lack understanding of the craft or any of the causal relationships
involved. Instead, they do what they‘ve seen their masters doing in similar situations, as
a result of their long exposure to medicine as doctors‘ slaves. They simply dictate,
without any capacity to explain why it is that they prescribe what they do in this
circumstance:
This kind of doctor never gives any account of the particular
illness of the individual slave, or is prepared to listen to one; he
simply prescribes what he thinks best in the light of experience,
302
Ἆξ‘ νὖλ θαὶ ζπλλνεῖο ὅηη, δνύισλ θαὶ ἐιεπζέξσλ ὄλησλ ηλ θακλόλησλ ἐλ ηαῖο πόιεζη, ηνὺο κὲλ
δνύινπο ζρεδόλ ηη νἱ δνῦινη ηὰ πνιιὰ ἰαηξεύνπζηλ πεξηηξέρνληεο θαὶ ἐλ ηνῖο ἰαηξείνηο πεξηκέλνληεο,…
νἴρεηαη ἀπνπεδήζαο πξὸο ἄιινλ θάκλνληα νἰθέηελ, θαὶ ῥᾳζηώλελ νὕησ ηῶ δεζπόηῃ παξαζθεπάδεη ηλ
θακλόλησλ ηῆο ἐπηκειείαο. 303
Jouanna observes that medicine in the Hippocratic corpus does not seem to be divided this way, though
he does use this reference to slave doctors to help explain the mysterious ―assistants‖ that appear in many
of the Hippocratic texts (Jacques Jouanna, Hippocrates [Baltimore: Johns Hopkins University Press, 1999],
90-91, 400).
143
as if he had precise knowledge, and with the self-confidence of a
dictator.(720c)304
The phrase for experience here is the term ―empeiria‖, a term encountered before in the
Gorgias. That dialogue was discussed in Chapter Two, in the discussion of why crafts
must aim toward a good and why pastrybaking is not a craft but merely a practice
(―empeiria‖). Slave doctors are similar. First, slave doctors lack any comprehensive
knowledge understanding of health and also any comprehensive understanding of the
causal relationships between the treatment prescribed and the health of the patient. As a
result, what slave doctors have to work with, which may very well work in many cases, is
having seen treatments and having seen illnesses. From this experience, they can
prescribe treatments they have seen doctors prescribe on the basis of the similarity of the
conditions of their patients to those they have seen doctors treat. However, slave doctors
cannot explain what they are doing or why they are doing it because they simply don‘t
understand what they are doing or why they are doing it. Slave doctors lack the
understanding of the workings of the body or the workings of the medicine.305
On the other hand, the Athenian visitor introduces what he calls the ―free doctor‖,
who proceeds in a quite different method than these slave doctors. His patients, unlike
those of the free doctor, are usually freemen. Because he actually understands the body
and the reasons for which he has prescribed the course of treatment he has prescribed, he
discusses what it is that he knows with his patient:
The visits of the free doctor, by contrast, are mostly concerned
with treating the illnesses of free men; his method is to construct
an empirical case-history by consulting the invalid and his
friends; in this way he himself learns something from the sick
304
θαὶ νὔηε ηηλὰ ιόγνλ ἑθάζηνπ πέξη λνζήκαηνο ἑθάζηνπ ηλ νἰθεηλ νὐδεὶο ηλ ηνηνύησλ ἰαηξλ δίδσζηλ
νὐδ‘ ἀπνδέρεηαη, πξνζηάμαο δ‘ αὐηῶ ηὰ δόμαληα ἐμ ἐκπεηξίαο, ὡο ἀθξηβο εἰδώο, θαζάπεξ ηύξαλλνο
αὐζαδο. 305
Aristotle makes a similar distinction between craft and experience in Metaphysics A:1. There, he argues
that experience often works very well at curing illnesses, but that it is ultimately distinct from craft because
it lacks understanding of causation: ―With a view to action experience seems in no respect inferior to
art...But yet we think that knowledge and understanding belong to art rather than to experience, and we
suppose artists to be wiser than men of experience...and this because the former know the cause, but the
latter do not‖ (―πξὸο κὲλ νὖλ ηὸ πξάηηεηλ ἐκπεηξία ηέρλεο νὐδὲλ δνθεῖ δηαθέξεηλ,...ἀιι‘ ὅκσο ηό γε εἰδέλαη
θαὶ ηὸ ἐπαΐεηλ ηῇ ηέρλῃ ηῆο ἐκπεηξίαο ὑπάξρεηλ νἰόκεζα κᾶιινλ, θαὶ ζνθσηέξνπο ηνὺο ηερλίηαο ηλ
ἐκπείξσλ ὑπνιακβάλνκελ...ηνῦην δ‘ ὅηη νἱ κὲλ ηὴλ αἰηίαλ ἴζαζηλ νἱ δ‘ νὔ‖) (981a13-b27).
144
and at the same time he gives the individual patient all the
instruction he can. (720d)306
This passage provides an important record of how free patients were treated, or at least
how Plato believes free patients ought to be treated.307
The remainder of this chapter will
focus on of what kind of instruction Plato means.
Notice that this passage treats the communication between doctor and patient as
largely two-way. On the one hand, the free doctor takes the time to compile an empirical
case-history using the information provided not only from the patient, but also his or her
friends. However, at the same time, he tries to educate the patient, in so far as he can.
Interestingly, the Athenian stranger says that this happens ―at the same time‖. Somehow,
the doctor explains what it is that he is trying to teach while he collects the information
from his patient. One minimalist interpretation of what is happening here might be that
somehow the instruction of the patient is a by-product of the collection of information,
not a deliberate attempt of the physician to instruct the patient. However, this
interpretation cannot be correct. Notice that the doctor only instructs the patient, not the
patient‘s friends. If the instruction were merely a by-product of collecting information,
then the patient‘s friends would be instructed as well. They are not. Instead, in the
306
ὁ δὲ ἐιεύζεξνο ὡο ἐπὶ ηὸ πιεῖζηνλ ηὰ ηλ ἐιεπζέξσλ λνζήκαηα ζεξαπεύεη ηε θαὶ ἐπηζθνπεῖ, θαὶ ηαῦηα
ἐμεηάδσλ ἀπ‘ ἀξρῆο θαὶ θαηὰ θύζηλ, ηῶ θάκλνληη θνηλνύκελνο αὐηῶ ηε θαὶ ηνῖο θίινηο, ἅκα κὲλ αὐηὸο
καλζάλεη ηη παξὰ ηλ λνζνύλησλ, ἅκα δὲ θαὶ θαζ‘ ὅζνλ νἷόο ηέ ἐζηηλ, δηδάζθεη ηὸλ ἀζζελνῦληα αὐηόλ. 307
The interpretation of this passage intersects with interpretations of the purpose of the prefaces to the
laws in the Laws as a whole, as the medical example serves as an analogy or an example by which one
might understand the purpose of legal prefaces. The Athenian stranger explicitly says that educating
patients includes the purpose of making them ―hēmeros‖, meaning ―tame‖ or ―civilized‖. Some have taken
this to imply that somehow the education is disingenuous. See, for instance, Christopher Bobonich, Plato's
Utopia Recast (Oxford: Oxford University Press, 2002), 97-105. Another example is Plato, Laws X:
Translated With a Text and Commentary, trans. Robert Mayhew (Oxford: Clarendon Press, 2008), 55-58.
However, the word ―tame‖ does not apply to humans as the word ―hēmeros‖ does. When one uses the
word ―tame‖ (a strictly beast-related word) for humans, it has the effect of appearing de-humanizing and
makes the entire process of education described in the Laws seem disingenuous. However, the word
―hēmeros‖, while its first meaning in the Greek-English Lexicon is the beast-related ―tame‖, also has other
meanings such as ―civilized‖ for humans, ―cultivated‖ for plants and ―ready for planting‖ for land (H.G.
Liddell and R. Scott, Greek-English Lexicon with a Revised Supplement [Oxford: Clarendon Press, 1996],
771). Given that the Greek ―hēmeros‘ has these secondary meanings that the English ―tame‖ does not, one
should not automatically assume that Plato intended to use the term in a dehumanizing way. Compare
―train‖. One may ―train‖ a doctor and ―train‖ a dog, but that does not imply that one treats a doctor like a
dog. The same term may apply to different classes without implying comparison of the members of those
classes. Plato even couples being ―hēmeros‖ with being most ―godlike‖ (―ζεηόηαηνλ κεξώηαηόλ ηε‖) at
Laws 766a. Therefore, I take the claims about teaching being appropriate to the free patient at face value.
145
process of collecting the information from the patient, the physician somehow overtly
makes an effort to instruct the patient ―all…he can‖.
In order to understand the content of the information that is appropriate to teach a
free patient, one must look to other passages in Plato. Later on in the Laws, in Book IX,
the Athenian stranger provides some more content to what the free doctor teaches to the
free patient:
The doctor would be acting almost like a philosopher, engaging
in a discussion that ranged over the source of the disease and
pushed the inquiry back to the whole nature of the body.
(857d)308
This provides some additional content to the instruction that was lacking in Book IV.
First, note that the free doctor here is said to be behaving, ―almost like a philosopher‖ and
is ―engaging in a discussion‖ with the patient. This hearkens back to the simultaneity of
collecting information and educating the patient that was mentioned in Book IV. The
free doctor discusses the various parts of the empirical case study with the patient as he
develops that case history. They discuss two other important things. First, they discuss
the source (―archē‖) of the disease. In other words, what they discuss is the causal
relationships involved in the disease which the physician has. Moreover, they discuss
this with respect to the ―whole nature of the body‖. This hearkens strongly back to the
discussion of dialectic in the Phaedrus that I discussed in detail in Chapter One and
Principle #4: Crafts are dependant on dialectic for a determinate subject matter and
to systematically approach causal relationships. What distinguishes a craftsperson
from someone with a mere empeiria is that the craftsperson has understanding of the
various techniques of medicine in relationship to the whole.309
The physician not only
explains the disease and the causal processes involved in the disease, but also health and
the body as a whole, providing the patient with the understanding of the goal of medicine
as well as the techniques of medicine or the aetiology of his or her own disease.
Medicine requires such individualized care because of the imprecision of the
craft. In fact, medicine is so complex that it is not subject to codification or to any rule-
308
θαὶ ηνῦ θηινζνθεῖλ ἐγγὺο ρξώκελνλ κὲλ ηνῖο ιόγνηο, ἐμ ἀξρῆο ηε ἁπηόκελνλ ηνῦ λνζήκαηνο, πεξὶ
θύζεσο πάζεο ἐπαληόληα ηῆο ηλ ζσκάησλ. 309
If this passage is connected to the Phaedrus, as it likely is, it would provide further evidence as to what
the reference to ―whole‖ in the Phaedrus means, that is, that it refers to the whole subject. However, the
word for ―whole‖ in this passage is ―pasēs‖, not ―holon‖.
146
like empeiria, like the one used by slave doctors. Rather, each case must be examined
particularly and in relation to the causal processes of the individual body. In the
Philebus, Socrates divides up crafts according to their level of mathematical precision:
SOCRATES: So there is a lot of imprecision mixed up in it and
very little reliability.
PROTARCHUS: Very true.
SOCRATES: And will we not discover that medicine, agriculture,
navigation, and strategy are in the same position? (56a-b)310
Medicine is less precise than some other crafts such as carpentry, and there will be more
imprecision in each medical decision. As a result, medicine is stochastic; each case must
be treated individually. This imprecision manifests itself in the rushed method of
performing slave medicine. Since medicine cannot be easily reduced to rules that are
simply transmitted, both the authors of the Hippocratic corpus and Plato recommend
individualistic treatment and the development of case histories. I have referred to
Hippocrates‘ method of observation over a long period of time in Section IXa, above.
The Athenian Stranger recommends a similar mode of treatment. Since medicine is a
stochastic and less precise craft, it cannot be reduced to simple rules. Rather, the causal
relationships both of the body and of the particular case must be understood, and related
to the particularities of a given case.
It is due to this imprecision that slave doctors end up in the position of dictators.
Because of the stochastic nature of the medical craft, the kind of associative rules of
thumb that one can generate through experience are insufficient and end up being applied
in inappropriate circumstances. When one does not understand the complexities of a
craft or the causal relationships involved, one must often create general rules that apply in
most cases. However, the slave doctor will not be able to know when exceptions arise,
since he or she does not understand the rules in the first place. Therefore, it is not enough
to simply know that something is usually beneficial. One must understand why
310
{Ω.} Οὐθνῦλ κεζηὴ κέλ πνπ κνπζηθὴ πξηνλ, ηὸ ζύκθσλνλ ἁξκόηηνπζα νὐ κέηξῳ ἀιιὰ κειέηεο
ζηνραζκῶ, θαὶ ζύκπαζα αὐηῆο αὐιεηηθή, ηὸ κέηξνλ ἑθάζηεο ρνξδῆο ηῶ ζηνράδεζζαη θεξνκέλεο
ζεξεύνπζα, ὥζηε πνιὺ κεκεηγκέλνλ ἔρεηλ ηὸ κὴ ζαθέο, ζκηθξὸλ δὲ ηὸ βέβαηνλ.
{ΠΡΩ.} Ἀιεζέζηαηα.
{Ω.} Καὶ κὴλ ἰαηξηθήλ ηε θαὶ γεσξγίαλ θαὶ θπβεξλεηηθὴλ θαὶ ζηξαηεγηθὴλ ὡζαύησο εὑξήζνκελ
ἐρνύζαο.
147
something is usually beneficial in order to recognise when it will not be. In Statesman,
the Eleatic visitor describes the absurdity of trying to perform medicine by using rules:
VISITOR: Are we to say - that is, between us- that if a doctor, or
else some gymnastic trainer, were going to be out of the country
and away from his charges for what he thought would be a long
time, and thought that the people being trained, or his patients,
would not remember the instructions he had given them, he
would want to write down reminders for them - or what are we
to say?
YOUNG SOCRATES: As you suggested.
VISITOR: But what if he came back unexpectedly, having been
away for less time than he thought he would be? Do you think
he wouldn‘t propose other prescriptions, contrary to the ones he
had written down, when things turned out to be different, and
better, for his patients because of winds or else some other of the
things that come down from Zeus which had come about
contrary to expectation, in some way differently from the usual
pattern? (295c-e)311
Even a doctor cannot render reliable the kind of empirical rules by which the slave doctor
tries to function. Understanding of what to do in a particular situation, then, is not simply
a matter of knowing what usually works. Such a rule requires understanding why
something usually works due to the causal principles underlying it. Once those principles
are understood, one will no longer need the rule, but will understand, in each case, why
something works.
What the free doctor provides, then, is not simply a set of rules for a patient about
how to be healthy, like the ―four food groups‖ or ―reduce salt‖. While these things
usually generate health, they do not carry with them the kind of detailed understanding of
health in the patient’s case, nor do they carry with them the understanding of the nature
of the body as a whole. What is described by the Athenian Stranger, then, in both Books
IV and IX of the Laws is an attempt to instruct the patient as far as possible in the
detailed causal history of his or her own illness and how it and the cure relate to the body
311
{ΞΔ.} Σλ ηνηλδε. εἴπσκελ γὰξ δὴ πξόο γε κᾶο αὐηνὺο ἰαηξὸλ κέιινληα ἠ θαί ηηλα γπκλαζηηθὸλ
ἀπνδεκεῖλ θαὶ ἀπέζεζζαη ηλ ζεξαπεπνκέλσλ ζπρλόλ, ὡο νἴνηην, ρξόλνλ, κὴ κλεκνλεύζεηλ νἰεζέληα ηὰ
πξνζηαρζέληα ηνὺο γπκλαδνκέλνπο ἠ ηνὺο θάκλνληαο, ὑπνκλήκαηα γξάθεηλ ἂλ ἐζέιεηλ αὐηνῖο, ἠ πο;
{ΝΔ. Ω.} Οὕησο.
{ΞΔ.} Σί δ‘ εἰ παξὰ δόμαλ ἐιάηησ ρξόλνλ ἀπνδεκήζαο ἔιζνη πάιηλ; ἆξ‘ νὐθ ἂλ παξ‘ ἐθεῖλα ηὰ γξάκκαηα
ηνικήζεηελ ἄιι‘ ὑπνζέζζαη, ζπκβαηλόλησλ ἄιισλ βειηηόλσλ ηνῖο θάκλνπζη δηὰ πλεύκαηα ἢ ηη θαὶ ἄιιν
παξὰ ηὴλ ἐιπίδα ηλ ἐθ Γηὸο ἑηέξσο πσο ηλ εἰσζόησλ γελόκελα.
148
as a whole, something that should be possible given time according to Principle #3:
Crafts are teachable. Only this sort of medicine is appropriate to a free person. To
provide simple rules without this understanding would be to treat the free patient like a
slave patient, and it is not a co-incidence that the Athenian stranger describes the slave
doctor as behaving with the confidence of a tyrant.
In fact, the Athenian Stranger claims that prescriptions given as rules without
understanding are both the province of slave medicine and are tyrannical (―tyrannos‖)
(720c). This is not because he is using rules per se, but because he is transmitting those
rules without any attempt to help the patient understand what he is doing. On the other
hand, as knowledge is the criterion for freedom, explaining what the physician is
planning to do and why he is planning to do it is the type of medicine appropriate to a
free patient. A physician who does understand the causal relationships and how they
contribute to the whole health of the body should take care to explain those causes as
much as possible.
8 Chapter Conclusion
Plato develops an understanding of the relationship between force, persuasion,
knowledge and freedom. Faced with the disintegration of the traditional synthesis that
allied persuasion with freedom and slavery with force, and the challenge presented to that
synthesis by rhetors like the historical Gorgias who tried to equate force and persuasion,
Plato takes neither side. Instead, he makes a distinction in persuasion between sorcerous
and didactic persuasion. The former relies on the manipulation of the emotions of the
hearer, but the latter involves providing the hearer with knowledge. This knowledge
becomes the new criterion by which someone can be considered free. In medicine, we
see a contrast between the sorcerous persuasion practiced by the character Gorgias and
the free persuasion practiced by the Athenian Visitor‘s free doctor. Given the stochastic
nature of the medical craft, the free doctor cannot simply prescribe rules. Instead, the
free doctor provides explanations to the patient that ―range over the source of the disease
and pushes the inquiry back to the whole nature of the body‖ (720c). This is a high
standard, but according to Plato, knowledge is what makes us immune to sorcerous
149
persuasion and is the source of freedom, and so knowledge on the part of the patient is
the only guarantee of, as we would now say, autonomy.
150
C h a p t e r 4 : M o d e r n A p p l i c a t i o n s
1 Introduction
Just as the Platonic understanding of crafts (―technai‖) proved fruitful in the
ancient world in debates about the nature of virtue, so too can it prove fruitful in modern
debates in medical ethics. Plato‘s claims about the relationships between types of
knowledge, between knowledge and power and between tool and product can all be used
to clarify and hone modern discussions. In this chapter, I will use what I will call Plato‘s
―framework‖ of crafts to shed further light on these discussions. My aim is not to
discover what Plato ―would have said‖ on any of these topics; far from it. Rather, I will
mine Plato‘s framework for distinctions and definitions that I will apply to modern
debates, clarifying some of the confusions and proposing approaches to the problems in
Platonic terms that, while they may not always resolve the debates, may enable them to
be discussed more fruitfully.
I will apply Platonic principles to three subjects, all related to Plato‘s hierarchy of
crafts that I discussed in Chapter One. First, I will discuss medical paternalism and
autonomy. Modern debates concerning paternalism and autonomy tend to assume that
the physician knows what is best for the patient‘s health, whereas autonomy only comes
into play as a kind of veto power in case the patient has other, non-health goals. I will
argue that, when one applies a Platonic framework and regards the patient as the expert in
the use of his or her own body, this simple paradigm breaks down. Instead, just like the
flautist directing the flute-maker in Republic X, it is the patient that is in the best position
to discern the relative value of the health goals weighed in any medical decision that
involves trade-offs. Second, I will use the Platonic description of the Hippocratic method
in the Phaedrus to discuss the use of recommendations in the ―Author‘s Discussions‖
sections of modern medical meta-analyses. The Phaedrus makes a careful distinction
between the understanding of the subject of the craft, causal relationships relevant to that
subject, and the application of those causal relationships to the subject. The methods
applied in meta-analyses can only provide causal understanding of techniques relative to
predefined indications and contraindications, while recommendations require
151
understanding of the importance of those aspects to the health of the patient as a whole.
To provide recommendations as though they are the scientific conclusion of a meta-
analysis fails to appreciate the limitations of the methodology and presupposes a scale of
value that, when it does not consider the patient‘s own scale of value, is paternalistic.
Finally, I will apply the Platonic framework of crafts to the current debate about the
―therapeutic misconception‖ in research ethics. Several modern bioethicists argue that,
because therapeutic medicine and medical research are separate crafts, they therefore
ought to have separate ethical systems. I argue that these bioethicists have only
appreciated a part of the relationship between therapeutic medicine and medical research.
Though they are separate crafts, they are hierarchically related; medical research
produces tools for therapeutic medicine. The physicians engaged in medical research
serve the role of evaluating the products of medical research, in a similar way to any user
evaluating a potential tool. Physician researchers are therefore still functioning qua
physicians, and therefore will continue to be bound by the rules of medical ethics despite
the distinction between the two activities.
2 Paternalism and Autonomy
I will first apply Plato‘s hierarchy of crafts to the problem of paternalism and
autonomy in medical decisions. What I will argue is that the current paradigm of
paternalism and autonomy is incomplete. In the current paradigm, the physician is
considered to be the expert in understanding what decision is best for the patient‘s health,
while the patient is the expert on how best to integrate health with other goals he or she
may have. Should the patient be able to choose between health and other goals, the
patient is autonomous, meaning that the patient is able to act on the basis of his or her
own beliefs about the good life. Should the physician try to hinder the patient‘s other
goals in favour of health, the physician is acting paternalistically, meaning that the
physician is trying to circumvent the patient‘s own vision of the good life in favour of
health, putatively for the patient‘s own best interests. The problem with this paradigm is
that the assumption that physicians know what is best for a patient‘s health is in itself
paternalistic. Except in trivial cases in which one negative outcome includes all other
152
negative outcomes, there is no single scale intrinsic to medicine for comparing competing
health goals. Arbitration between competing health goals requires understanding of the
uses to which the patient wishes to put the body, and claims by physicians to know which
health goal is best would require knowledge of what way of life is best. As a result, in
order to understand the proper relationship between physician and patient, I will apply
Plato‘s arguments about the relationships between tool producer and tool user.
Specifically, Plato argues that crafts that produce tools are not complete. They need
supervision by those intending to use the tool to ensure that what is produced is useful
and to evaluate the kinds of trade-offs that need to be made in production. The body is a
tool with which the patient may live the kind of life he or she wishes to live. The patient,
therefore, is required to compare competing health goals, not to simply compare health to
other goods in order to be autonomous.
2.1 The Current Paradigm
The current paradigm for considering paternalism and autonomy includes the claim
that the physician knows what is best for the health of the patient, while the patient, then,
can only contribute other, non-health goals that he or she may have.312
Paternalism
occurs when a physician tries to unduly influence a patient to seek health over other
goals. The patient‘s autonomy, therefore, can only interfere with medical practice, and
the patient‘s input is reduced to a kind of veto of health in favour of other goals. This
paradigm assumes four things, all of which I will argue are incorrect:
A) There is always a single outcome that can be considered best for the health of the
patient.
B) The physician is the expert on both the identification and production of this
outcome.
C) Paternalism occurs only when a physician unduly promotes health against non-
health goals.
312
Note that the paradigm I refer to here is not a complete paradigm for paternalism, but a common
approach to the division of knowledge between physician and patient that occurs within that paradigm.
153
D) A patient‘s autonomy is limited to weighing health against non-health goals.
As one might expect, most contemporary discussions of paternalism argue that
paternalism should be avoided. This is a laudable goal. However, those same
discussions make these same assumptions about the kind of expertise that the physician
brings to a health care decision, and about the purely negative role that autonomy plays in
a medical decision. In this section, I will go through contemporary texts on paternalism
to show how these three assumptions are routinely made.
The first example is from Informed Consent: Legal Theory and Clinical Practice.
In that book, Jessica W. Berg, Paul S. Appelbaum, Lisa S. Parker and Charles W. Lidz
summarise the origin of the desire for patient autonomy in the following way: ―What
individual patients asserted against medical paternalism was, in essence, that health was
not the only value of importance to them.‖313
Note how this passage is phrased. Health
is treated as a single value. Berg et al. do not say that autonomy arose as the result of a
patient‘s desire to choose which health objective is of most value to them, but only of
other values versus this single value of health. Patient autonomy is limited to comparing
the value of health against other, non-health goals the patient might have in an overall
vision of the good life.
In the book Autonomy & Paternalism: Reflections on the Theory and Practice of
Health Care, Thomas Nys et al. make a similar claim. They argue that health is an
―agent-neutral good‖, meaning that, all things considered, it is something that all people
would agree is a good thing:
Even if we count it [health] as an intrinsic good, it is not the only
intrinsic good nor the summum bonum. The science and the
practice aimed at producing health and overall functioning then
is an ‗agent-neutral good‘...This medical good is an agent-neutral
good because it depends neither on an individual physician‘s
vision of a good more broadly conceived, nor on that of a patient
or the patient‘s family or social context. 314
Note how Thomas Nys et al. make the same assumptions as those listed above. Like
Berg et al, also that Nys et al. speak of health in the singular. It is only a single goal at
313
Jessica W. Berg, Paul S., Parker, Lisa S. Appelbaum and Charles W. Lidz, Informed Consent: Legal
Theory and Clinical Practice (New York: Oxford University Press, 1987), 20. 314
Thomas Nys, Yvonne Denier and Toon Vandevelde, Autonomy & Paternalism: Reflection on the Theory
and Practice of Health Care (Leuven: Peters, 2007), 60-61.
154
which we may aim. The term ―overall functioning‖ implies that there is some measure
against which we may gauge what, if there are health outcomes to weigh, would best
contribute to our health ―overall‖, and that knowledge of this measure is a part of the
physician‘s expertise. This good is something that exists independently of any particular
broader views of a good life, and would be the same for each patient regardless of his or
her life plans. Then, in order to argue against paternalism, they argue that this unified
health is not the only good. Our autonomy only allows us to weigh health against the
other goals that we might have. Nys et al. use the contemporary paradigm: health is a
unified goal, medicine‘s aim is that goal and autonomy‘s only capacity is to assert other
goals against health.
Note what this paradigm implies. Whenever a patient dissents from a physician‘s
recommendation, the patient is considered to be choosing some other non-health goal
over his or her own health. One might expect that, when there is a trade-off between one
health outcome and another, and a patient chooses to go against a physician‘s advice so
as to pursue one of those outcomes, it would be characterized as being the choice of one
health goal over another. However, on this paradigm, anyone who prefers one health
outcome over another contrary to a physician‘s advice is considered to be acting on non-
health goals. There is only one outcome that is best relative to the ―agent-neutral‖ goal
of health, and a patient who chooses another health outcome is considered to be acting for
the sake of some other goal. What is best for the body is known by the physician
irrespective of the uses we ourselves might have for it, rendering irrelevant any individual
life goals to the very practice that repairs our bodies. This makes individual beliefs about
the good life, the capacity of whose pursuit constitutes our autonomy, irrelevant to
health-care decisions per se.
2.2 The Powers of the Body
In this section, I will discuss the powers of the body, using the analogy of other
complex tools that we use.315
However, though I will be adopting a Platonic approach, I
315
One difficulty in describing the body as a tool is that it seems to rule out simply enjoying bodily activity
for its own sake. This difficulty raises issues that there is insufficient space to address thoroughly. In brief,
―tool‖ need not imply more than that having something in a certain condition makes some activity possible.
155
will need to argue paradoxically that Plato‘s model of the relationship between tools,
tool-makers and tool-users was incomplete. Though Plato mentions the relationship
between tool-maker and tool-user in both the Cratylus and Republic X, he never provides
a systematic account of the collaboration between tool-user and tool-maker. This lack of
a systematic account may be the result of a limitation of Plato‘s account of the excellence
of tools, especially in Republic I. There, Plato treats tools as being evaluated (indeed,
being defined) by a single characteristic power, rather than by a collection of powers. In
this section, I will discuss tool examples that do not conform to Plato‘s one-tool, one-
power analysis, and then discuss those examples with relation to the body. I will argue
that the powers within a single tool conform to certain logical relations that affect the
way that they interact.
According to Plato, each tool has a function or ―ergon‖.316
Good tools have the
power to perform their ergon well. They have this power by virtue of some excellence or
―aretē‖. For example, a pruning knife has the function of cutting (its ergon). It is a good
knife when it has the power (dynamis) to cut vines well. It has the power to cut vines
well by virtue of its sharpness (its aretē). The clearest example of this relationship is
presented in the Republic, Book I. First, he establishes the connection between the
function of something and the unique power of something: ―And would you define the
function of a horse or of anything else as that which one can do only with it or best with
it?‖ (352e).317
Note how he defines the ―function‖ in terms of what something can do
(elsewhere, ―dynamis‖, though power is indicated by a dative of means here). Some
single, unique power defines its function. Note the distinction between a function and a
power: the power is the ability to do something, while the function is actually doing it.
Plato further says that there is a single excellence related to each function. This
excellence is what allows the thing to perform the activity well.
For instance, I run ―with‖ my legs, though running is, at least in part, an activity of my legs. Even Plato,
the dualist, says that ―we sense with the soul‖ (―ηῇ ςπρῇ αἰζζαλόκεζα‖) (Theaetetus 185d3). Something
can use a part of itself if it performs some activity by taking advantage of some power of that part. This
allows one to make use of Plato‘s discussions of ―use‖ and ―tools‖ without necessarily accepting his
dualism. 316
I discussed the translation of ―ergon‖ in chapter one. ―Function‖ is the usual translation, but the Greek
word can apply to non-tools as well as tools. ―Characteristic activity‖ is better in some contexts. 317
Ἆξ‘ νὖλ ηνῦην ἂλ ζείεο θαὶ ἵππνπ θαὶ ἄιινπ ὁηνπνῦλ ἔξγνλ, ὃ ἂλ ἠ κόλῳ ἐθείλῳ πνηῇ ηηο ἠ ἄξηζηα;
156
Now, I think you‘ll understand what I was asking earlier when I
asked whether the function of each thing is what it alone can do
or what it does better than anything else…[I‘m asking] whether
anything that has a function performs it well by means of its own
peculiar excellence and badly by means of its badness. (353b)318
In the Republic I, then, Socrates provides a one-to-one-to-one relationship of function to
power to excellence. A single function is provided by a single power made possible by a
single excellence.
The difficulty with Plato‘s analysis is that it doesn‘t apply to many tools, even to
some of the tools that he himself lists. I will go through some of those counter-examples
to show the complexity. The complexity manifests itself in the following ways:
A) Some tools are capable of multiple functions (they have multiple erga).
B) Some tools have their function through a system of activities that can be
evaluated independently (they have multiple aretai).319
I will go through both of these conditions in turn, and then discuss the ways in which
these different complexities manifest themselves in the body, and how the body does not
have a one-to-one-to-one relationship of function to power to excellence.
The first example I will discuss is the example of tools that have multiple erga or
functions. Probably the clearest example of this is the Swiss Army Knife. Unlike its
pruning knife cousin, a Swiss Army knife is capable of numerous activities of different
kinds. It has appendages for cutting fruit, meat, opening beer bottles, opening wine
bottles, filing nails, cutting paper and turning screws, among other things. Unlike simple
tools, which have a single ergon, a Swiss Army knife has at least as many erga as it has
appendages, and perhaps more than that. If one considers the Swiss Army knife a single
tool, then it is a single tool with multiple erga, not a single ergon, and Plato‘s one-one-
one analysis would not apply.
One might say in response that a Swiss Army knife is not actually a single tool,
but rather multiple tools on a single base. A Swiss army knife‘s powers nicely correlate
to its physical parts. However, other tools are not so easily broken into parts. Take, for
318
Ἥηηο, ἤλ δ‘ ἐγώ, αὐηλ ἀξεηή· νὐ γάξ πσ ηνῦην ἐξση, ἀιι‘ εἰ ηῇ νἰθείᾳ κὲλ ἀξεηῇ ηὸ αὑηλ ἔξγνλ
εὖ ἐξγάζεηαη ηὰ ἐξγαδόκελα, θαθίᾳ δὲ θαθο. 319
I will discuss a third sort of complexity involving trade-offs and design restrictions in sub-section 2.4.
157
instance, a water-proof wind-breaker. People quite commonly wear such a windbreaker
for the sake of keeping out both the cold wind and for keeping off the wet rain (assuming
they aren‘t also using an umbrella). We need the water-proof windbreaker for both
functions. It is much harder to simply divide up the two functions of the water-proof
windbreaker into distinct physical parts the way one does a Swiss Army knife. Saying
the water-proof windbreaker is two tools is implausible unless one is willing to say that
the same parts of the same physical object can be two different tools. Further, even if one
were willing to say that the same physical object can be two different tools, Socrates‘
definition allows of an ergon allows for the possibility of something having multiple
erga. He says that a tool‘s ergon is something that it alone can do or that it can do better
than anything else. This is not inconsistent with a tool‘s being able to perform two erga
better than anything else. For instance, a hearth was (at the time) the best way of heating
a home and the best way of cooking food. Even on Socrates‘ definition, the hearth would
therefore have two erga.
The second type of complex tool I will discuss is tools that have a single ergon,
but produce this ergon through a system of sub-powers.320
As a consequence, these tools
have multiple aretai, one for each of the sub-powers. I will use Plato‘s own example of a
flute from Republic X. Let us say that the function of a flute is to produce pleasure
through melody. The problem is that the flute does this by having multiple parts that
each have their own distinct powers. The flute has a mouthpiece that produces the
original whistling sound when blown on, a shaft that textures the whistle that has been
made, and pips that must be properly proportioned so that one may produce mathematical
relationships among the notes to produce a melody. Note that there are three different
powers here that all go together to create a tool capable of producing pleasure through
melody. One part of the flute might be defective, while the other parts work very well.
320
This argument has some affinity with that in Paul Sheldon Davies, Norms of Nature: Naturalism and the
Nature of Functions (Cambridge: MIT Press, 2003). There, he argues that function should always be
defined hierarchically in the sense that a function isn‘t a function unless it contributes to some further
activity that could not function without it. For instance, the function or functions of the nervous system can
be defined relative to locomotion or the manipulation of objects. However, while Davies‘ argument allows
for the recognition of discrete powers as parts of larger functional systems, it creates the difficulty that the
final erga, that is, powers of the body that don‘t contribute to larger physical systems are not functions. My
argument corrects this omission by distinguishing between final erga, like those of a Swiss army knife and
contributing powers, like those of the flute.
158
For instance, the reed in the flute might need replacing,321
but the pips and the shaft are in
―working order‖. Because a flute is a system of powers, each of those powers has its own
excellence, and so the flute (or at least the parts of the flute) may be said to have multiple
excellences.
One might object that the virtue of the flute should be considered all of these parts
working together well, so the parts of the flute can‘t be called excellences on their own.
The problem with this objection is that it takes away some of the linguistic space we need
to discuss repairing objects. If I take my car in to the shop because there is something
wrong, and I ask the mechanic, ―Is there something wrong with the carburetor?‖, he or
she might answer, ―No, the carburetor is working. The ignition is broken‖. If we deny
that parts can be excellent independently of one another, this discussion becomes
impossible. My carburetor still has the power to regulate the amount of gasoline and air
going into an engine, which is its own ergon. The carburetor has this excellence
independently of the excellences of the other parts. In fact, it even has this excellence
when sitting on the shelf before it is even placed in a car. I do not mean to imply here
that the excellence of carburators can be defined independently of the excellence of cars.
What I am arguing is that the excellence of a given carburetor can be evaluated
independently of a given car, even the car in which it is currently placed (and even when
it is not placed in a car at all). When the overall activity of something follows from a
sequence of activities of the parts, the excellences of the parts can be evaluated
independently.
The human body has features of both sorts of complexity: it has multiple erga,
like a Swiss army knife, and it has multiple systems with multiple excellences, like a
flute. First, the body is like a Swiss army knife in that there are different things we can
do with the different parts of the body, and in some cases different things we can do with
the same parts of the body. To provide the most obvious examples, we see with our eyes,
we hear with our ears and we move. Some parts of our body have quite different powers:
we eat, speak and breathe with our mouths, while we touch, manipulate objects and catch
ourselves from falling with our hands. These are different erga: there are multiple things
321
Since the flute is a Platonic example, I am speaking here of the Greek aulos, usually translated ―flute‖.
It is a reed instrument. It is more like a modern oboe than a modern flute, but I use the translation ―flute‖
as it is the traditional one.
159
that we can do with our bodies, none of which is reducible to another.322
In this respect,
then, the human body deviates from the one-one-one analysis of a pruning knife in favour
of the multiple powers of a Swiss army knife.
Second, the body is also in many ways like a flute, in that it has a number of parts
that each contributes to the various erga of which the body is capable. The body has a
number of contributing and overlapping systems that interact in ways analogous to and in
some ways more complex than the parts of the flute. Consider, for example, an act of
walking. In order to walk, one must have energy that is provided through nutrition, a
respiratory system for oxygen, a nervous system to send information, muscles and
ligaments for the movement of the legs, and bones to support the body. This act of
walking would be of greater or lesser quality depending on the quality of the sensory
powers involved. Sight can make it easier to walk, though it is not necessary. Each of
these sub-systems has excellences that can be evaluated independently of one another,
much like the carburetor and the ignition. Nutrition and respiration are required for all
erga; they are ―vital‖ systems. Damage to the nervous system can affect multiple
systems depending on the location of the damage. Finally, some systems can contribute
to other systems without being necessary. The body combines both the complexities of
the Swiss-army knives and the flute. It has multiple erga, but also has a complex
arrangement of sub-powers relative to those erga. What this means is that the body can
be said both to have multiple functions (erga) and multiple excellences (aretai).
2.3 Healthier and Unhealthier
In this section, I will use the above discussion of complex powers to challenge the
first assumption of the current paradigm, that there is a single medical outcome that can
be said to be best for the health of the patient. Even if there is such a thing as a perfectly
healthy body, as Normal Daniels says when he describes health as ―the natural functional
organization of a typical member of a species‖,323
it would not imply a single scale
322
A misunderstanding of evolution might lead to the conclusion that the purpose of our bodies is
reproduction, but this rests on a confusion between the origin of a power and the ergon of a power. 323
Daniels‘ complete definition of health is ―the absence of disease and diseases...are deviations from the
natural functional orgainization of a typical member of a species‖ (Norman Daniels, "Health Care Needs
160
against which to compare deviation from that natural functional organization. That one
can describe a perfectly functioning Swiss-army knife does not imply a single scale
against which one can say which broken attachment would make the Swiss-army knife
more dysfunctional ―overall‖. What I will argue is that when a tool has multiple erga,
there is no internal scale against which to compare malfunctions in those erga. By an
internal scale, I mean a scale that can be derived from understanding the erga themselves.
Instead, when a tool has multiple erga, an external scale must be used to compare the
relative value of those erga.324
By an internal scale, I am referring to a scale defined solely according to the
compared powers themselves. For example, in comparing sight, a scale might be the
20/20 vision scale, while in comparing mobility, a scale might be speed. To compare
sight against hearing, however, some third, external scale would need to be used to
compare the two. Only in trivial cases in which only one power is impeded or in which
one negative outcome includes another negative outcome can one sick person be said
with certainty to be healthier than another using internal scales. For example, all other
things being equal, someone who is blind is less healthy than someone who is sighted and
someone who is blind and deaf is less healthy than someone who is only blind. Between
powers, however, no internal scale exists against which to measure one deviation from
another. There is no internal scale of deviation by which one might compare sight versus
mobility, or sensitivity in the limbs versus ease of digestion. The current paradigm would
suggest, in Nys et al.‘s terms, that there is a single, internal scale called ―overall
functioning‖ to which health contributes. Ashcroft and McMillan refer to our ―medical
welfare‖, and say that knowledge of the patient‘s medical welfare is in the hands of the
physician. This suggests a single internal scale of deviation that can be applied to
compare any two unhealthy people. However, nothing is said about how this is to be
measured. The presence of multiple erga makes such a unity as difficult to gauge as it
would be to gauge the lack of which attachment would most affect the overall functioning
and Distributive Justice", Philosophy and Public Affairs, 10 [1981]: 156). I have removed the unnecessary
double negative. 324
This is true even of more complex systems than a Swiss Army knife. Powers that have no immediate
use, such as circulation, are useful exactly in so far as they contribute to those that do have such use. While
complex systems might give the appearance that they contribute to some sort of overall scale of
functioning, they in fact are only evaluated in terms of the sum of the powers for which they are necessary.
161
of a Swiss Army knife. Any scale of deviation would need to come from outside a metric
of the powers themselves, as the scale would require measuring the powers against one
another.
The more complex and interdependent the powers, the more complex the range of
malfunction, and the harder it is to compare one deviation from another. Even
apparently unified erga such as sight are far more complex than they might appear. For
instance, sight can be affected in many different ways. All things being equal, someone
with 20/30 vision is healthier than someone with 20/40 vision, because those two powers
are being compared against a single, measurable scale. However, one person might be
near-sighted (measured on the 20/x scale), while another person might be farsighted,
unable to focus on nearer objects, but with 20/20 vision. Another person might be
colour-blind, unable to distinguish between red and green, while another might be colour-
blind, unable to distinguish between blue and yellow. A fifth person might have floaters
and cataracts, obstructing portions of the field of vision, while a sixth person might have
only one, perfectly functioning eye, but with no depth perception and a limited field of
vision on one side. A person might even have different defects in different eyes. Our
language of ―20/20 vision‖ obscures the different kinds of defects that people have, such
that even the phrase ―better vision‖ has no single internal scale against which vision
might be compared.325
What is true for vision is even more true for the body as a whole.
It should be noted that I am not denying either Plato‘s claim that there is some
overall healthy state of the body of which the physician has knowledge, nor am I denying
Normal Daniels‘ claim that health can be defined as normal species functioning.326
Both
of these claims can be retained. If one conceives of health as the normal range of powers
of a human being, then one can conceive of what it would mean for a person to be
325
This claim about the complexity of diseases is very similar to Aristotle‘s claim that disease is multiform
(―polueides‖). That there are so many ways to deviate from a single form leads to a great deal of variation. 326
This distinction is important in order to avoid some of the difficulties introduced by a position like that
of Christopher Boorse‘s distinction between disease and illness. He argues that disease should be
considered deviation from natural species functioning, while illness is defined of any bodily state that
makes us unhappy. The difficulty with such a distinction is that the severance is too extreme as it renders
what he calls ―disease‖ irrelevant to medical decisions, and takes away any standard by which we might
differentiate disease from lacking some sort of non-natural bodily enhancement (Christopher Boorse, "On
the Distinction Between Disease and Illness", Philosophy and Public Affairs, 5 [1975]: 49-69).
162
perfectly healthy.327
All of the parts of the perfectly healthy person‘s body would
function in line with some sort of species-typical norms. Knowledge of these species-
typical norms would be a part of the medical craft. In fact, they must be a part of the
medical craft, or there would be no goal at which medicine would aim. Nonetheless, the
presence of species-typical norms does not provide an internal scale against which to
compare defects so as to say that one ill person is more ill than another. In the case of
any proposed treatment involving trade-offs that there is no single internal scale against
which to compare the alternatives. Medicine per se has a single goal, health, but that no
more implies that there is an internal scale than that the intelligibility of a perfectly
functioning Swiss Army knife implies there is an internal scale against which its
attachments may be weighed.328
I will also answer a possible objection to this position. One might argue that it is
common sense that the ability to walk is more important than the ability to touch one‘s
thumb to one‘s little finger, and that therefore the physician qua physician should be able
to say that a person with paraplegia is less healthy than a person with a damaged joint in
the little finger. Therefore, my claim that medicine cannot internally compare
malfunctions is epistemically implausible. My response is that even to make this
evaluation requires an external scale, in this case, common sense. That the capacity to
walk is more important than the capacity to touch one‘s thumb and little finger cannot be
derived internally from the definitions of the powers themselves. In this case, it is true
that any person would prefer to be able to walk than to touch together one‘s thumb and
little finger, but that requires the application of an external scale. That every practically
327
This is similar to what Robert Wachbroit calls a ―biological concept‖ of normality, which is an
understanding of normality rooted in the normal biological functioning of various organs working together.
This allows for the consideration of ―perfect health‖ without necessarily believing that there is anyone who
is perfectly healthy (Robert Wachbroit, "Normality as a Biological Concept", Philosophy of Science, 1994:
579-91). 328
A special comment should be made about the vital powers, those powers without which one would be
dead. These powers are special in terms of comparing health and disease because, from the point of view
of functioning, death is worse than any other possible diseased state. Death includes all other possible
dysfunction. Therefore, from the point of view of function, death is the worst possible outcome. In effect,
the vital powers provide a special example of a trivial case: just as being blind and paralysed is unhealthier
than just being blind, being dead is unhealthier than any diseased state. However, this special status only
applies to certain death, and only applies to the cessation of a vital function, not its impairment.
Diminishment of a vital power does not have the same special logical status as its absence, because its
diminishment is does not necessarily include the destruction of other powers. For example, difficulty in
digestion doesn't affect sight. Therefore, when it comes to evaluating who is healthier and who is
unhealthier, vital powers have a special logical status, but this logical status only extends to their absence.
163
rational human being, including every physician qua practically rational human being,
would make such a decision does not make the scale internal, only a matter of universal
intuitive consensus. In the next section, I will consider those cases where common sense
does not provide an adequate scale of malfunction, and another external scale must be
applied. Nonetheless, whether one refers to common sense or another scale, when there
are multiple erga, the scale must be external.
The multiple erga of the human body imply that, except in trivial cases,
comparing health states is impossible from within the craft of medicine itself. Such
comparisons would require a single metric. These comparisons are impossible internally
even if one accepts the claims of Plato and Norman Daniels that there is a definite,
perfectly healthy state of the body. Even knowledge of such a healthy state does not
supply a single metric for comparing deviations so long as that perfectly healthy state
includes multiple erga.
2.4 Medical Decisions and Collaboration
In the previous section, I dealt with what it meant to say that someone is healthier
or unhealthier than another person, and the problems that having multiple erga presents
for thinking about this. In this section, I will discuss the applications of this to medical
decisions themselves. Often, in health care decisions, trade-offs must be made between
various health goals. The expertise required to make decisions when trade-offs must be
made lies not in the internal expertise of medicine, but in the external expertise of
understanding the uses to which the body will be put. In other words, the best metric for
making a medical decision comes from the craft of living, which the patient is presumed
to have.329
In order to explain this position, I will use the example of a shield-maker and
329
In saying that the patient is presumed to have the craft of living, I deviate somewhat from Plato‘s own
beliefs, in that he rather clearly did not believe that everyone had the craft of living, and assumed that such
a craft was possessed only by a few people. My goal is to reconcile, in so far as is possible, Plato‘s theory
with liberal democracy. However, I have deliberately chosen the term ―presumed‖ rather than ―assumed‖.
One can support liberal democratic theory without necessarily assuming that everyone knows how best to
live. Rather, one presumes, for practical purposes, that people know best how to live their own lives so as
to promote autonomy and to preserve the maximum scope for practical reason. In other words, even if one
does not believe that someone has the craft of living, one acts as though that person does, much in the way
that a defendant is presumed innocent until proven guilty, even when the evidence is overwhelming. One
can thereby agree with Plato that there is a craft of living without necessarily drawing the same political
164
a soldier. I will also add a distinction not made explicit in Plato. This is a distinction
between the excellence of something relative to its powers and the excellence of
something relative to its use. Using these distinctions, I will show how the patient is a
required part of any medical decision and is needed in order to supply medicine with its
completion.
Consider two shields. The first is a ten-foot shield that can fully conceal the body
but is impossible to carry. The second is a three-foot shield that can both be carried and
used effectively with one‘s weapon of choice. Which is the more excellent shield?330
This question raises an overall tension in Plato‘s thought about tools. According to the
passage in Republic I, it would seem that the ten-foot shield is actually a better shield. If
one defines shield strictly as ―projectile blocker‖, then the ten-foot shield is a better
projectile blocker and therefore a better shield. However, in Republic X and the
Cratylus, the story becomes more complicated. The craftsperson who is to use the shield
needs to direct its production because of the restrictions placed on design by its use: ―It‘s
wholly necessary, therefore, that a user of each thing has most experience of it and that he
tell a maker which of his products performs well or badly in actual use‖ (Republic
X:601d).331
Note that a good tool here is being defined not in terms of some one capacity
that it has (as in Book I), but in terms of its ―actual use‖, which admits of other
considerations. Plato here admits a kind of complexity that was not in Book I.
Specifically, he admits that considerations of use may be admitted into defining quality,
above and beyond the power of the tool. Excellence in tools is being defined relative to
the using craft, not relative to a defining power.
conclusions. Another option would be to take the Protagorean route of arguing that most people do, in fact,
have the craft of living, and therefore it should be assumed in patients rather than simply presumed. In
either case, presuming that the patient has the craft of living is a deviation from Plato‘s own theory, which
puts such a craft in the hands of the political class, specifically the guardians. 330
A case like this is considered by Plato in Timaeus 75b-c. There, the demiurge makes the human skull,
but needs to trade-off two different capacities. On the one hand, the skull needs to be thin enough to allow
for perception. On the other hand, it needs to be thick enough to protect the brain. If one tries to define the
skull in terms of one or the other of these powers, it would be less excellent. It‘s only when one balances
the two powers in actual use that the chosen thickness of the skull can be said to be ideal. 331
Πνιιὴ ἄξα ἀλάγθε ηὸλ ρξώκελνλ ἑθάζηῳ ἐκπεηξόηαηόλ ηε εἶλαη θαὶ ἄγγεινλ γίγλεζζαη ηῶ πνηεηῇ νἷα
ἀγαζὰ ἠ θαθὰ πνηεῖ ἐλ ηῇ ρξείᾳ ᾧ ρξῆηαη.
165
One way to think of this is to think of the function in two different senses, the first
of which I will call its ―effect‖ and the second of which I will call its ―use‖.332
Take, for
instance, the example of a bridle. We tend to describe a bridle in terms of it being ―a tool
to steer a horse‖, and this is the correct description of it. Note, however, that this
describes the bridle in terms of its role in a higher-order craft, riding. Another way to
describe a bridle is ―a tool that turns a horse‘s head‖. It would be quite possible to use a
bridle to turn a horse‘s head without the purpose of riding it, defining it in terms of a
single effect. Defining something‘s excellence in terms of its use requires the
contribution of those who will actually use the object, because the use of the object refers
explicitly to the activity of the user. Creating a bridle, defined as a tool for moving a
horse‘s head, does not necessarily require the input of a rider. A bridle, defined as a tool
for steering a horse, would require the supervision of a horse rider. The same issue arises
with the warrior and the shield. When a shield is described as a tool for ―blocking
projectiles‖, it does not take into account its actual use. However, when it is described as
―for blocking weapons in battle‖ or ―for blocking blades by an infantryman‖, then that
description includes its use. Similarly, a body, defined as a tool for ―pumping blood,
moving around and moving stuff‖ does not necessarily require the input of a patient to
understand these effects, but a body, defined as a tool for ―living a life of sort x‖ does
require the input of the patient to understand what sort of effects are most useful to living
that sort of life.
When a tool is defined in terms of its actual use, rather than simply in terms of its
effects, the user‘s input provides the external scale for evaluating the relative value of
erga. In medical decisions, this becomes important whenever trade-offs between erga
must be made. I will provide an example. A man has a non-malignant tumour on his
spine between the twelfth and thirteenth thoracic vertebrae (about two-thirds of the way
down the spine). The presence of the tumour is causing pain to the patient that is
inhibiting, but not preventing mobility. The physician believes she can remove the
tumour, but removing it would carry a risk of causing irreparable and permanent paralysis
in approximately 5% of cases. Should the patient undergo the procedure? Here we have
332
Plato does not make this distinction. Nonethess, this distinction can help sort our some of the puzzles
involved in understanding collaboration between tool-maker and tool-user, as well as trade-offs made in the
production of certain tools.
166
a case where the restoration of complete function must be weighed against a chance of its
complete loss. However, there is no internal scale of deviation that can weigh these two
possible outcomes against one another. Rather, the patient, who intends to use the body,
must weigh the possible outcomes and their probabilities against the kind of life he
wishes to live. Depending on the patient‘s career and leisure activities, a partial
inhibition of mobility may be more or less of a concern. If a patient has a career that
involves a great deal of lifting, or if the patient is a serious athlete, undergoing the
procedure may be a risk more worth undertaking. After all, in those cases, a serious
impairment might be just as devastating to a moving or sports career as complete
impairment. If the patient is less physically active, the partial mobility he or she has may
be more acceptable, and so the patient would be less willing to undergo the procedure, as
the risked full impairment would bring new problems with it of its own. In either case,
because there is not an internal scale against which to compare the outcomes against one
another, one must use an external scale, in this case, the life goals of the patient.333
A quick word should be said about risks to vital powers, where one might, for
example, have a risk to an ergon on one side, with a risk to life on the other. Even more
complex choices are possible, such as a choice between a heart operation that would
create a moderate chance of immediate death, but, unless it is undertaken, there would be
a high risk of death in a few years. In cases like this, the patient is choosing different
levels of probability for different lengths of lives based on the goals that the patient has
for his or her life overall, and the length of time and strength of body needed to reach
those. A person whose vision of the good (remainder of) life involves spending as much
time as possible with family may think differently about a risk of death from someone
whose vision of a good life includes regular physical activity as a necessary component.
Even when risks of death are involved, what kind of life one wishes to live are important
to evaluating those risks.
333
For such a comparison to be made, it is not necessary that the various powers be commensurate (that
they can be measured using a single unit), but comparable (that they can be ranked). For some choices, it
might seem that no comparison is possible, as the chooser would be unable to make a judgment as to which
is better. However, not being able to judge which option is better does not necessarily imply that they are
incomparable. Rather, they may be comparable, and have been judged to be ―roughly equal‖. This latter
point was made in Thomas Hurka, Virtue, Vice and Value (New York: Oxford University Press, 2001),
176-77.
167
What all of this implies is that the patient needs to be involved in any medical
decision that involves any trade-offs among powers.334
Absent an internal standard in
medicine by which to compare the health outcomes of various medical options, the
patient is required to evaluate which outcomes are more useful. The body is something
that will be used, and there are no internal metrics between powers that can say which
power will be of more use to a given patient in any actual life. Instead, a process of
collaboration between physician and patient is necessary. The physician knows which
erga are actually damaged and various procedures that may or may not be able to restore
or improve those powers at which rates of success. The physician also knows the various
risks of those procedures and which erga may be adversely affected by those risks. The
patient, on the other hand, has the understanding of what kind of life he or she wishes to
live, and is able to consider how those various powers will contribute or not to that vision
and to what degree. Both are necessary to a medical decision, as the goal of the medical
decision is to provide a body fit for use by the patient.
2.5 Rethinking the Paradigm
The current paradigm of considering the physician as the expert on what is best
for the patient‘s health, while the patient can only contribute non-health goals, appears on
its face to combat paternalism and promote autonomy. However, it ultimately does not
go far enough and builds paternalistic elements into itself. By considering the
relationship between erga and excellences with the body, one can see that there is no
internal scale in medicine to determine between competing health outcomes. What this
implies is that the putative expertise of the physician qua physician to know what one
outcome is best for a patient‘s health does not exist, except in trivial cases. Believing that
it does, however, has two negative side-effects. First, it means that any patient who
chooses a health outcome that the physician does not believe is the best outcome is cast
as someone who is choosing to act against his or her own health for the sake of some
334
This is not to say that all medical decisions are controversial. Many tradeoffs will be obvious given
common sense. However, this common sense is not possessed by the physician qua physician but qua
human being with common sense about the good life. In effect, it is possessed qua potential patient, in
much the same way as a shield maker who also happens to be a soldier.
168
other goal. Let us take the case of the man who must choose whether or not to undergo
back surgery for a non-malignant tumor. If the physician advises that he undergo the
procedure, but he doesn‘t wish to undergo it, the current paradigm would suggest that this
patient is sacrificing his health for the sake of other, non-health goals. However, the
proper characterization is that the patient is not willing to risk full immobility for the sake
of removing an impediment to immobility. Adding this extra element in which a patient
is somehow sacrificing ―health‖ whenever he or she prioritizes his or her erga in a way
different from a physician is an inaccurate and unfair characterization of what he or she is
doing.
Secondly, the current paradigm actually promotes paternalism, though indirectly.
The paradigm promotes paternalism because it promotes the physician‘s view of the good
life over that over the patient‘s, putatively in the name of the patient‘s best interest.
Since there is no internal way of ranking erga, the physicians must be getting their beliefs
about which powers are more important from somewhere. In order to rank the powers of
the body, then, the physician must be presupposing some view of what constitutes a good
life and building that view into his or her recommendations. When a physician makes a
recommendation based on views about the good life, whether those of her own or of a
community, without allowing for possible contrary views by the patient, this is
paternalistic. It hampers the ability of the patient to live based on his or her own view of
the good life in favour of the physician‘s own view of what is in the patient‘s own best
interests. However, at least if it were phrased in terms of different views of the good life,
the patient might be in a position to recognize value differences and choose differently.
When these views are couched in a technical guise, masquerading as professional
expertise, the patient may not have enough information to recognize the conflict, and, as
stated above, any deviation is portrayed as sacrificing health. The current paradigm
thereby allows a great deal of paternalism to slip through unnoticed.
Instead, my suggested paradigm builds patient autonomy right into medical
decisions. Rather than seeing the patient‘s view of the good life as some external
constraint on medicine, the patient is actually required for making correct medical
decisions, because the body is a tool for that very patient‘s use. The patient is just as
essential for making medical decisions as a soldier is for making shields, and a patient‘s
169
input is no more a constraint on medicine than a soldier‘s input is a constraint on shield-
making. The patient‘s view of the good life becomes a part of every medical decision,
because it serves as the goal of the medical decisions and the only metric against which
medical decisions can be weighed. I therefore suggest that the paradigm should be
revised. The physician is not the expert who knows what is best for the patient‘s overall
health while the patient is only the expert on his or her other goals. Rather, medicine is a
lower-order craft, repairing and preserving the tool we all use for living. This will
remove the subtle paternalism of the current paradigm and improve patient autonomy.
3 Recommendations in Meta-Analyses
3.1 Introduction
In this section, I will combine some of the principles of the ―Hippocratic Method‖
from the Phaedrus with some of the considerations from the last section. I will then
apply those concepts to medical meta-analyses. A meta-analysis is a statistical
aggregation of studies in an attempt to compile all the available data on a given question.
The meta-analyses employ selection criteria for which studies will be included. Usually,
these studies have a section called ―Author‘s Conclusions‖ that provide a
recommendation. I will argue in this section that these authors' conclusions typically are
not, and in fact cannot be, entailed by the data that is provided in the meta-analysis. This
is the result of two problems with the method as a whole. First, the methods employed in
meta-analysis can only be applied to causal relationships. Second, any recommendation
would include presuppositions about the relative priority of various health outcomes in
patients, something that cannot be determined by the studies and that introduces the
difficulties from Section I. I will show that these studies routinely confuse the causal
conclusions of the study with the weighing of outcomes, and I will suggest that these
studies not include recommendations.
170
3.2 The “Hippocratic Method” and the Scientific Method
In the Phaedrus, Socrates makes a distinction between the three parts of what he
there calls the ―Hippocratic Method‖. This method consists of three distinct steps. The
first involves a process of collection and division, in which one recognizes what comes
together in the same kind, and then the various parts within that kind. The second step
involves understanding the various parts and the powers they have, defined as how they
affect other things and, more importantly, how they are affected by other things. The
final step applies these causal relationships so as to produce the desired state of the
subject studied in the first step. I discussed that method in Chapter One, but in this
chapter, I will focus on this distinction between the three parts and why it is significant
for understanding the method of meta-analyses. Specifically, meta-analyses can only
include the second part of the Hippocratic Method, not the first or third part of the
Hippocratic Method. The statistical analysis they perform can only apply to causal
relations between properly defined indications, but cannot contribute to the proper
understanding of the proper functioning of the body nor the importance of those
indications.
The three parts of the Hippocratic method are described more than once by
Socrates in the Phaedrus, and are taken as an analogy from medicine and applied to
rhetoric. What he calls the ―Method of Hippocrates‖ is divided into three distinct parts.
When applied to the soul, Socrates describes it in the following terms:
SOCRATES: First, you must know the truth concerning everything
you are speaking or writing about; you must learn how to define
each thing in itself; and, having defined it, you must know how
to divide it into kinds until you reach something indivisible.
Second, you must understand the nature of the soul, along the
same lines; you must understand which kind of speech is
appropriate to each kind of soul, prepare and arrange your
speech accordingly, and offer a complex and elaborate speech to
a complex soul and a simple speech to a simple one Then, and
only then, will you be able to use speech artfully, to the extent
that its nature allows it to be used that way, either in order to
teach or in order to persuade. (277b-c)335
335
Πξὶλ ἄλ ηηο ηό ηε ἀιεζὲο ἑθάζησλ εἰδῇ πέξη ὧλ ιέγεη ἠ γξάθεη, θαη‘ αὐηό ηε πᾶλ ὁξίδεζζαη δπλαηὸο
γέλεηαη, ὁξηζάκελόο ηε πάιηλ θαη‘ εἴδε κέρξη ηνῦ ἀηκήηνπ ηέκλεηλ ἐπηζηεζῇ, πεξί ηε ςπρῆο θύζεσο δηηδὼλ
171
The method has three parts: application of dialectic to the subject so as to understand it as
a whole and to divide it into kinds (or parts);336
the discovery of causal relations; and the
application of those causal relations to the whole. These steps are interrelated. The first
step provides understanding of the whole subject and its genē (types or parts). These
genē are needed to ask the appropriate causal questions of the second part of the method,
and those causes are then used to produce the desired state of the whole subject,
understood dialectically from the first part.
This distinction in the Hippocratic Method has importance for understanding the
methods of meta-analyses. The use of statistical methods in medical practice must
necessarily be divided into three parts. First, before a particular study can take place, the
researcher must already have a clear question to pose. A medical study looks for certain
indications or contraindications, measurable objectives or risks that a physician would
seek or avoid. Not all indications or contraindications can be examined in any given
study, so there is already some selection of the relative value and disvalue of these
indications and contraindications in this selection process. Second, the study must
examine the causal relations between proposed treatments and those selected indications
and contraindications using statistical analysis.337
Finally, physicians using these
analyses must apply the discovered causal relationships to the body. If there is some
trade-off between indications or between indications and contraindications, the physician
must also have in mind some external scale for ranking the various indications and
contraindications. Note, however, that the statistical methods of a meta-analysis can only
contribute to the second part of this three-step process. A meta-analysis cannot get
started unless one has presupposed the indications and contraindications that will be
searched for. Further, a meta-analysis cannot be applied unless one has an external scale
against which to weigh any possible trade-offs between the indications and
contraindications.
θαηὰ ηαὐηά, ηὸ πξνζαξκόηηνλ ἑθάζηῃ θύζεη εἶδνο ἀλεπξίζθσλ, νὕησ ηηζῇ θαὶ δηαθνζκῇ ηὸλ ιόγνλ, πνηθίιῃ
κὲλ πνηθίινπο ςπρῇ θαὶ παλαξκνλίνπο δηδνὺο ιόγνπο, ἁπινῦο δὲ ἁπιῇ, νὐ πξόηεξνλ δπλαηὸλ ηέρλῃ ἔζεζζαη
θαζ‘ ὅζνλ πέθπθε κεηαρεηξηζζῆλαη ηὸ ιόγσλ γέλνο, νὔηε ηη πξὸο ηὸ δηδάμαη νὔηε ηη πξὸο ηὸ πεῖζαη. 336
The term kinds (―γέλε‖) here refers to both types and parts in various sections of the Phaedrus. I
discuss the reasons for the double use in Chapter One. 337
Strictly speaking, statistical methods can observe only correlation, not causation. However, the
inferences from these correlations are causal and their applications require that they be such. Part of the
role of control groups in trials is to establish the direction of the causation.
172
Only the discovery of causal relations can be, strictly speaking, a part of a meta-
analysis. First, one must choose certain measurable indications and contraindications for
which to search. Second, the causal relations must be established. Finally, one must
decide how to apply these causal relations in actual practice. The statistical analysis of
meta-analyses can only contribute to the second step in this process. The indications and
contraindications themselves, as well as any scale on which to balance trade-offs are
necessarily external to the discovery of causal relationships.
3.3 Actual Cases
In this section, I will show through examples that, despite these methodological
limitations, medical meta-analyses claim to be able to perform the third step in the
―Hippocratic Method" when they claim that they have actually demonstrated which
course of treatment is better using the analysis. As meta-analyses are limited at most to
showing causal relationships, this is an error, but it is a common and even ubiquitous one.
For this section, I have chosen the January 2009 issue of The Cochrane Review, the most
prestigious meta-analysis journal. In nine studies within that one issue, authors either
claim they have weighed or that they will weigh risks against benefits, as though that is
something that is within the scope of their methodology. They do this in two sections: at
the beginning, when they set out what it is that they intend to prove; and at the end, when
they summarise their conclusions.
Meta-analyses in the Cochrane Review routinely claim that their purpose is to
weigh one risk against another in a way that involves comparative evaluation, rather than
simply establish causal relationships through empirical data. The claim often appears
early in the review, when the review justifies its importance. In the January 2009 issue,
seven reviews present the problem that they hope to address as being that of whether or
not harms outweigh benefits:338
Corticosteroids have powerful anti-inflammatory effects and
have been used to treat established CLD. However, it is unclear
338
The reason many of these claims are in the past tense even though they are about what the article will
demonstrate is that they are outlining the flaws in the current literature that they hope to correct.
173
whether any beneficial effects outweigh the adverse effects of
these drugs.339
After GOG #92 showed that adjuvant radiotherapy reduced the
number of recurrences, the debate has changed to whether this
benefit is enough to outweigh the attendant risks. To our
knowledge there has been no previous systematic review of this
subject.340
However, little evidence currently exists to demonstrate whether
there is any additional benefit in combining the two treatments
compared with either treatment alone and, if there is, that this
benefit outweighs any disadvantages or risks associated with the
additional therapy (Spiegel 1997; Wardle 1990; Westra 1998).341
This Cochrane systemic review aims to determine whether
nutritional supplementation before surgery is associated with
better outcomes than no supplementation and whether the
benefits of pre operative nutrition outweigh any potential harms
in improving clinical outcome of malnourished patients
undergoing abdominal surgery.342
On balance, however, potential benefits from statins appear to
outweigh potential detrimental effects and adverse effects from
statins will be assessed in this review.343
Even so, a key question to be answered is whether the potential
risks of the treatment are outweighed by their benefits in the
management of acute COPD exacerbations.344
Notice how each of these quotations present the problem that their meta-analysis is
intended to address: risks and benefits have not yet been weighed. They are not simply
claiming that they will be demonstrating causal or statistical relationships, which is the
limitation of their method. In these articles, confusion about the limitations of the
method of meta-analyses appears right from the beginning of the studies.
339
H. L. Halliday, R. A. Ehrenkranz and Doyle L. W., "Late (>7 Days) Postnatal Corticosteroids for
Chronic Lung Disease in Preterm Infants", Cochrane Database of Systematic Reviews, 2009: CD001145. 340
L. Rogers et al., "Adjuvant Radiotherapy and Chemoradiation after Surgery for Cervical Cancer",
Cochrane Database of Systematic Reviews, 2009: CD007583. 341
N. Watanabe, R. Churchill and T. A. Furukawa, "Combined Psychotherapy Plus Benzodiazepines for
Panic Disorder", Cochrane Database of Systematic Reviews, 2009: CD005335. 342
J. Khan and S. A. Razak, "Preoperative Nutrition to Prevent Post Operative Complications in
Malnourished Patients Undergoing Elective Abdominal Surgery", Cochrane Database of Systematic
Reviews, 2009: CD007597. 343
B. McGuinness, R. Bullock and D., Kerr, E., Passmore, P. Craig, "Statins For the Treatment of
Alzheimer‘s Disease and Dementia", Cochrane Database of Systematic Reviews, 2009: CD007514. 344
J. A. E. Walters et al., "Systemic Corticosteroids for Acute Exacerbations of Chronic Obstructive
Pulmonary Disease", Cochrane Database of Systematic Reviews, 2009: CD001288.
174
The same confusion appears at the end of the studies. Several reviews not only
claimed that their purpose is to have ―weighed‖ risks and benefits, but that their evidence
either had provided or might have provided just such a successful weighing.345
In some
cases, the authors conclude that they lacked the evidence to decide whether risks
outweighed benefits, as though this is something they could possibly have shown:
The benefits of early postnatal corticosteroid treatment (≤7
days), particularly dexamethasone, may not outweigh the known
or potential adverse effects of this treatment.346
The benefits of late corticosteroid therapy may not outweigh
actual or potential adverse effects.347
There is insufficient evidence to conclude that the clinical
benefits of TMLR outweigh the potential risks.348
Although the question remains whether the possible deleterious
effects of reversible rejection in a very low percentage of
patients outweigh the possible beneficial effects of steroid
avoidance, the vast majority of patients do benefit from being
without steroids early after transplantation without immediate
risk of rejection.349
It seems that CsA partially outweigh350
the benefits in lipid
profile seen after stopping of steroids.351
The more intense diabetogeneicity of TAC partially outweighed
the benefits of steroid-sparing strategies in NODAT incidence.352
345
In this subsection, I am only discussing the methodological limitations of these studies and how these
limitations are not well reflected in the conclusions. In some cases, one choice of treatment may
―outweigh‖ another with the application of common sense to the causal inferences. However, in the next
two subsections, I will discuss how these weighings and common sense tend to come apart in many cases,
and how this methodological confusion is not a mere mischief. 346
H. L. Halliday and R. A., Doyle, L. W. Ehrenkranz, "Early (< 8 days) Postnatal Corticosteroids for
Preventing Chronic Lung Disease in Preterm Infants.", Cochrane Database of Systematic Reviews, 2009:
CD001146. 347
H. L. Halliday, R. A. Ehrenkranz and Doyle L. W., "Late (>7 Days) Postnatal Corticosteroids for
Chronic Lung Disease in Preterm Infants", Cochrane Database of Systematic Reviews, 2009: CD001145. 348
E. Briones, J. R. Lacalle and I. Marin, "Transmyocardial Laser Revascularization Versus Medical
Therapy for Refractory Angina", Cochrane Database of Systematic Reviews, 2009: CD003712. 349
J. Pascual et al., "Steroid Avoidance or Withdrawal for Kidney Transplant Recipients.", Cochrane
Database of Systematic Reviews, 2009: CD005632. 350
The phrase ―partially outweigh‖ appears in a few studies. Since it is impossible to partially outweigh
something in a literal sense, the phrase indicates that ―outweigh‖ is sometimes being used in a non-literal
sense. In these cases, ―outweigh‖ seems to being used synonymously with ―offset‖. 351
J. Pascual et al., "Steroid Avoidance or Withdrawal for Kidney Transplant Recipients.", Cochrane
Database of Systematic Reviews, 2009: CD005632. 352
ibid.
175
Budesonide may confer benefit in terms of a lower mean CDAI
score and a longer time to relapse of disease. However these
mild benefits are outweighed by the risk of adrenocorticoid
suppression when using budesonide for extended periods of
time.353
The potential adverse effects of statins among people at low risk
of CVD are poorly reported and unclear (Jackson 2001) but,
among Statins for the primary prevention of cardiovascular
disease those with pre-existing CVD, the evidence suggests that
any possible hazards are far outweighed by the benefits of
treatment.354
Both the introductions and the conclusions of these studies discuss weighing as though it
is both a purpose and an outcome of the studies themselves. However, such a weighing
cannot be performed by the methods of the studies, as they are limited to only the second
part of the Hippocratic Method. Because they use statistical methods to study causation,
they are able to determine whether or not there is causation in many cases. In fact, the
breadth of their sources makes them often the best source for discovering causal relations
between agents and the body. Nonetheless, the authors repeatedly claim that that is not
all they are doing. This is the result of believing that the method of meta-analyses can
somehow weigh outcomes, even though their methods are limited to causal relationships.
None of these meta-analyses provide a clear scale against which they are
weighing the benefits and risks of the various treatments they are comparing. However,
even if they were to provide such a scale, that scale would be subject to the same
criticisms I discussed in Section Two. Specifically, in order to perform the kind of
weighing that they propose to do and to have done in the meta-analyses, there must be a
single scale on which to do such weighing. However, this raises both problems raised in
Sections 2.3 and 2.4.355
First, no single, internal metric exists because of the multiple
erga of the human body. Second, any putative scale would necessarily contain
presuppositions about the good life that individual patients may or may not share.
Therefore, the inclusion of such scales necessarily produces the same kind of paternalism
353
E. I. Benchimol et al., "Budesonide Formaintenance of Remission in Crohn‘s Disease", Cochrane
Database of Systematic Reviews, 2009: CD002913. 354
F. Taylor et al., "Statins for the Primary Prevention of Cardiovascular Disease", Cochrane Database of
Systematic Reviews, 2009: CD004816. 355
As in those sections, there would be cases where common sense would dictate that one outcome should
be chosen over another. Nonetheless, this common sense would still be an external scale. For some cases
where common sense seems to break down in these cases, see the next section.
176
that the current paradigm of paternalism does. Specifically, presuppositions about the
good life are built into the scale and belong either to the medical community or to the
individual authors of the studies. These presuppositions take the form of paternalism
when the authors claim to have discovered the best course of treatment and thereby put
the patient in the position of believing that dissent is not dissent about what constitutes
the best state of the body relative to their life goals, but the choice of non-health goals
over the health goals presented by the authors of the study.
3.4 Concern about Precision
Another difficulty with the mixing of the third part of the Hippocratic Method
with the second is that it undermines the overall precision of meta-analyses. One of the
strengths of meta-analyses is their ability to aggregate data from hundreds of different
studies into a single study that is able to use statistical methods to discover causal
relationships. Through this model, they are able to discover statistical correlations
between the various treatments and medical indicators of a power that is simply
unmatched. They are also very good at noting when data is insufficient, and pointing out
where further research is required. However, when adding the first and third parts of the
Hippocratic Method, they become extremely imprecise. The scales of goals are
undefined, and the applications of the causal relations to those goals are equally unclear
and rarely stated. This has the effect of creating a strange hybrid of precision and
imprecision in a single document. It has the further effect of masking the imprecision of
the recommendations behind the precision of the results, creating the impression that the
conclusion has been as precisely demonstrated as the statistical analysis. In this section, I
will discuss my study of the ways in which the recommendations of meta-analyses are
confused with statistical results not by the authors themselves, but by those who cite the
meta-analyses. The goal is to show that the mixture of precision and imprecision creates
a false sense of precision about the conclusions.
In order to examine how meta-analyses were cited, I chose the 2006 meta-
analysis, ―Induction of labour for improving birth outcomes for women at or beyond
177
term‖ by A.M. Gülmezoglu, C.A. Crowther and P. Middleton.356,357
I examined ten
papers that cited this study to examine if any distinction was made between the causal
relations and the conclusions.358
The results were disappointing. Of those ten references,
not one even quoted the meta-analysis verbatim. Moreover, not one of the references
cited a page number. Instead, various bits of the meta-analysis were paraphrased and
brought together. It would be impossible without having read the original study to
determine to what extent they were citing the body of the meta-analysis proper and to
what extent they were citing the author‘s conclusions. The causal relations and the
practical conclusions had simply been mixed together. Two of the citations, those of
Stotland et al. and Heimstad et al., did appear to be citing the author‘s conclusions as
conclusions, though they actually made the conclusions sound more stringent than they
were. The original study made the following recommendation in its ―Author‘s
Conclusions‖: ―Labour induction at 41 completed weeks should be offered to low-risk
women.‖. Note the use of the term ―offered‖. Both of the above mentioned reports,
however, paraphrase this recommendation not as ―offered‖ but as ―policy‖. Stotland et
al. make the following statement, with the Cochrane review as its citation:
Multiple reports have suggested that a policy of induction of
labor at 41 instead of 42 weeks‘ gestation does not increase the
356
A. M. Gülmezoglu, C. A. Crowther and P. Middleton, "Induction of Labour for Improving Birth
Outcomes for Women at or Beyond Term", Cochrane Database of Systematic Reviews, 2006: CD004945. 357
The reason for choosing this older study, rather than one from the 2009 issue, is that it has had time to
be cited by secondary authors. 358
The ten sources were: A. B. Caughey, J. M. Nicholson and A. E. Washington, "First- vs Second-
Trimester Ultrasound: The Effect on Pregnancy Dating and Perinatal Outcomes", American Journal of
Obstetrics and Gynecology, 2008: 703.e1-6; J Vellekoop et al., "Indications and Results of Labour
Induction in Nulliparous Women: An Interview Among Obstetricians, Residents and Clinical Midwives",
European Journal of Obstetrics and Gynecology, 146 (2009): 156-9; H. Gatward et al., "Women‘s
Experiences of Being Induced for Post-Date Pregnancy", Women and Birth, 23 (2009): 3; J. M. Nicholson
et al., "Active Management of Risk in Nulliparous Pregnancy at Term: Association Between a Higher
Preventive Labor Induction Rate and Improved Birth Outcomes", American Journal of Obstetrics and
Gynecology, 2009: 254e1-254e13; N. E. Stotland, A. E. Washington and A. B. Caughey, "Prepregnancy
Body Mass Index and the Length of Gestation at Term", American Journal of Obstetrics and Gynecology,
2007: 378.e1-5; A. B. Caughey et al., "Who is at Risk for Prolonged and Postterm Pregnancy?", American
Journal of Obstetrics and Gynecology, 2009: 683.e1-5., Maria L. Costa et al., "Audit and Feedback: Effects
on Professional Obstetrical Practice and Healthcare Outcomes in a University Hospital", Acta Obstetricia et
Gynecologica Scandinavica, 88 (2009): 793-800; E. Mozurkewich et al., "Indications for Induction of
Labour: A Best-Evidence Review", British Journal of Obstetrics and Gynecology, 2009: 626-36; Ulla-Britt
Wennerholm et al., "Induction of Labor Versus Expectant Management for Post-Date Pregnancy: Is there
Sufficient Evidence for a Change in Clinical Practice?", Acta Obstetrica et Gynecologica Scandinavica, 88
(2009): 6-17; and, Runa Heimstad, Pål R. Romunstad and Kjell Å. Salvesen, "Induction of Labour for Post-
Term Pregnancy and Risk Estimates for Intrauterine and Perinatal Death", Acta Obstetricia et
Gynecologica Scandinavica, 87 (2007): 247-9.
178
cesarean rate and may reduce maternal and infant morbidity.
(683.e4)359
Heimstad et al. make the following statement, with the Cochrane review as its citation:
―A policy of induction of labour at 41 completed weeks reduces this risk without
increasing the risk of assisted vaginal or abdominal delivery‖.360
These sentences are
references as though the policies suggested are from the Cochrane Review. However, in
both cases, the ―offer‖ recommended in the Cochrane Review has turned into ―policy‖.
The lack of precision has been noted by those who author meta-analyses as well.
Olsen et al. in the paper, ―Quality of Cochrane Reviews: Assessment of Sample from
1998‖,361
argue that the treatment recommendations at the end of Cochrane Reviews are
rarely supported by the data. According to Olsen et al., of the 53 reviews in the fourth
issue of 1998, 17% of the ―Author‘s Conclusions‖ are were not supported by the data,
according to the eleven methodologists they consulted. In order to claim this, Olsen et al.
share some of the confusion of the Cochrane Reviews themselves, in that they believe
that there even might be a conclusion that would be supported by the data; I have
discussed that difficulty above. Olsen et al. do not provide a hypothesis as to why this
disconnect between the conclusions and the data occurs, but this disconnect can be
understood with reference to the Hippocratic method. The authors of the study have no
way of garnering a practical conclusion from their data directly. They therefore cannot
derive such a conclusion without importing assumptions about the best sort of life,
assumptions that may vary from person to person. It is not surprising, then, that a third
party reading a meta-analysis would disagree with the conclusions, as they may not
import the same assumptions about the good life.362
359
N. E. Stotland, A. E. Washington and A. B. Caughey, "Prepregnancy Body Mass Index and the Length
of Gestation at Term", American Journal of Obstetrics and Gynecology, 2007: 683.e4. Stotland et al. here
cite four different studies. However, the Cochrane Review says nothing about policy and shouldn‘t be cited
in support of this sentence. 360
Runa Heimstad, Pål R. Romunstad and Kjell Å. Salvesen, "Induction of Labour for Post-Term
Pregnancy and Risk Estimates for Intrauterine and Perinatal Death", Acta Obstetricia et Gynecologica
Scandinavica, 87 (2007): 248. 361
O., Middleton, P., Ezzo, J. Olsen et al., "Quality of Cochrane Reviews: Assessment of Sample from
1998", British Medical Journal, 2001: 829-32. 362
One thing that this study shows is that what is thought to be common sense can itself be controversial.
Presumably, the authors of the studies cited by Olsen et al. thought that they were applying common sense
to arrive at their conclusions. Nonetheless, what is thought to be common sense seems to arrive at
controversial conclusions at least a sixth of the time.
179
There is a real imbalance between the precision of the statistical analysis and the
imprecision of the authors‘ conclusions. This hybridization of the studies leads to
imprecision in the way that the analyses are used in actual practice. Rather than citing
the statistical results of the study, authors cite the conclusions, and even then, not
necessarily accurately. Moreover, they cite these conclusions as though they have the
full rigour of the statistical analysis itself.
3.5 Conclusion
The use of recommendations in meta-analyses is creating serious confusion
between the three parts of what Socrates called the ―Hippocratic Method‖ in the
Phaedrus. The scientific method is able to show causal relations using measurement and
statistical analysis. It is not, however, able to make recommendations or ―weigh‖ various
risk factors against each other. To do this requires knowledge of the relationship of these
risk factors to the whole, which is beyond the scope of the scientific experiments.
However, by including the recommendations at the end of these studies, the authors are
making claims for their studies that are far beyond what they have actually demonstrated
and undermine the scientific integrity of the paper. Moreover, confusion as to the
function of these conclusions allows them to both remain a part of the analysis for the
purpose of citation while simultaneously being insulated from the rigour and quality of
the rest of the analysis. As it is unlikely that these confusions can be easily sorted out, I
recommend that these ―Author‘s Conclusions‖ not be included in future meta-analyses.
4 The Therapeutic Misconception
4.1 Introduction
According to the Helsinki Declaration of 1964, medical research on human
subjects should be held to the same standards as therapeutic medicine, and all human
research subjects should ―be assured of the best proven diagnostic and therapeutic
180
method‖.363,364
However, authors such as Franklin Miller and Howard Brody have
charged that this conflation of ethical principles has led to confusion between the aims of
therapeutic medicine and medical research, generating what they call the ―therapeutic
misconception‖. Under this misconception, they say, physicians routinely compromise
the care of their patients on the grounds that their research is ―therapeutic‖, while patients
routinely assent to compromised care, unaware of possible conflicts between their
medical care and medical research. Therefore, they argue, therapeutic medicine and
medical research ought to have separate ethical codes and the medical research ethics
code need not guarantee the best proven therapeutic methods to their subjects. In this
section, I will argue that, while they are correct to note the important distinction between
therapeutic medicine and medical research, they are incorrect to conclude that this
distinction implies that medical research should be autonomous from the ethics of
therapeutic medicine. To demonstrate this, I will introduce the Platonic theory of the
hierarchy of crafts, in which Plato argues that crafts are arranged in a hierarchy in which
the work of tool producers must be evaluated by the tool users. Physicians participate in
trials not as researchers, but as physicians evaluating potential products of medical
research. Therefore, the ethics of research can never be as autonomous as Miller and
Brody would like.
4.2 The Therapeutic Misconception
In this section, I will apply the Platonic methods of dividing and ordering crafts to
a contemporary debate about the interaction of medicine and medical research.
Specifically, a challenge has recently been raised to the current ethical codes in medical
research. Since the Helsinki Declaration of 1964, the World Medical Association has
required that medical research may never compromise the health of the subjects involved
in a study: ―Concern for the interests of the subject must always prevail over the interests
363
World Medical Association, "World Medical Association Declaration of Helsinki: Ethical Principles for
Medical Research Involving Human Subjects", in World Medical Association Website
<http://www.wma.net/en/30publications/10policies/b3/index.html> [accessed 4 November 2009]. 364
There is some similarity between the principles of the Helsinki Declaration and the principle of ―never
doing harm‖ from the Crito 49b. The principle in the Crito is broader than that in the Helsinki Declaration,
but would itself justify such an approach.
181
of science and society.‖365
Franklin Miller and Howard Brody argue that the idea that
research in a therapeutic context will not interfere with patient care is a ―therapeutic
misconception‖. Instead, they argue that medicine and medical research are two different
activities with different methods and different ends that may conflict. To confuse
medicine and research causes numerous practical problems in the performance of both
and creates a serious threat to the autonomy of research subjects who are usually
mistaken as to the nature of the task into which they have entered. I will argue that those
who argue that there is a ―therapeutic misconception‖ are correct that there has been
confusion between medicine and medical research, but that they are mistaken in believing
that the logical distinction between medicine and medical research implies that there
ought to be separate ethical codes for each. In fact, the hierarchical relationship between
research and medicine implies that their ethical codes may never be completely severed.
In order to explain the origin of the therapeutic misconception, Miller and Brody
describe what they call the ―similarity position‖ and the ―difference position.‖366
According to the similarity position, medicine and research are the same activity and
therefore ought to have the same ethical codes. According to the difference position,
medicine and research are different activities and therefore ought to have different ethical
codes. Miller and Brody hold that the similarity position is incorrect and the difference
position is the correct one. However, if one subscribes to the difference position, there is
simply no need to reconcile the two ethical codes. For instance, one need not reconcile
the ethics of medicine, which require healing one‘s patients, with the ethics of law, which
require protecting a client‘s legal interest. They are different activities and therefore have
different ethical codes. Instead, they argue, ―The basic goal and nature of the activity
determines the ethical standards that ought to apply.‖367
In other words, the ethics of
crafts are entirely determined by the craft‘s nature and purpose and their ethics will differ
insofar as their nature and purpose differ. Since medical research and therapeutic
medicine have different goals and natures, they should therefore have different ethical
365
World Medical Association, "World Medical Association Declaration of Helsinki: Ethical Principles for
Medical Research Involving Human Subjects", in World Medical Association Website
<http://www.wma.net/en/30publications/10policies/b3/index.html> [accessed 4 November 2009]. 366
Franklin G. Miller and Howard Brody, "A Critique of Clinical Equipoise: Therapeutic Misconception
and the Ethics of Clinical Trials", The Hastings Center Report, 2003: 23. 367
Ibid., 22.
182
standards. Confusion over the natures and purposes of research and medicine is causing
those like the authors of the Helsinki Declaration to needlessly and confusingly conflate
the ethical codes of two separate activities.
In many ways, Miller and Brody are making a remarkably Platonic-sounding
argument.368
They have made a distinction between research and therapy that depends on
the distinction between their ends. Research has the goal of ―scientific knowledge‖,
while therapy has the goal of health. As a result, they claim, they are different activities.
This is the same kind of logical distinction made in the Republic and the Euthydemus: a
single craft has a single goal, and crafts ought to be distinguished by their goals.369
Miller and Brody also share the Platonic approach in treating these distinctions as
important, both logically and ethically. Socrates uses this ―strict speaking‖ about the
specificity of a craft to refute Thrasymachus in Republic I about the nature of justice, and
similar refutations occur in other dialogues. Confusion about what one is doing can have
serious ethical consequences. Miller and Brody argue that confusion about what
constitutes medicine and research leads to the systematic compromise of patient care in
ways that violate the integrity of physicians, the autonomy of patients and the practicality
of ethical codes. They believe that, should the distinction between research and medicine
be properly understood, these problems would be resolved.
From a Platonic perspective, research seeks to discover the causal relationships of
various potential treatments on the body so that medicine may then assign those
treatments for the sake of health. However, this implies that research and medicine are
distinct powers. Research is an acquisitive craft whose goal is to capture knowledge, in
much the same way as the crafts of geometry or arithmetic capture logical truths in
figures and in words. Medicine is a productive craft whose goal is the health of the
368
The argument also bears an affinity with Aristotle‘s distinction between theoretical and practical
knowledge from Nicomachean Ethics VI. However, Plato‘s framework of dividing crafts between the
acquisitive and productive is closer to Miller and Brody‘s treatment of research and therapy as being goal-
oriented ―activities‖. 369
For example, at Republic 346c, Socrates argues that ―We are agreed, then, that each craft brings its own
peculiar benefit‖ (―Οὐθνῦλ ηήλ γε ὠθειίαλ ἑθάζηεο ηῆο ηέρλεο ἰδίαλ ὡκνινγήζακελ εἶλαη;‖). At
Euthydemus 291e, Socrates uses several examples including medicine to show that crafts have a single
product: ―SOCRATES: Then what would you say its result was? For instance, if I should ask you what result
does medicine produce, when it rules over all the things under its control, would you not say that this result
was health‖ (―{Ω.} Σί νὖλ ἂλ θαίεο αὐηῆο ἔξγνλ εἶλαη; ὥζπεξ εἰ ζὲ ἐγὼ ἐξσηῴελ, πάλησλ ἄξρνπζα
ἰαηξηθὴ ὧλ ἄξρεη, ηί ἔξγνλ παξέρεηαη; νὐ ηὴλ ὑγίεηαλ <ἂλ> θαίεο;‖).
183
patient. They are logically distinct. Each has its own goal. Therefore, if one activity was
to also produce the goal of the other activity, this would be an accident, in the same way
that one does not pick carrots for the sake of health as the goal of carrot-picking is
carrots, and if one becomes healthy while carrot-picking, this is accidental.370
Crafts with
separate natures and purposes may conflict in their applications, as what leads to the good
of one subject may not lead to the good of another. In this way, a patient could be
harmed in the course of discovering a new therapy, while pulling a patient from a trial
may be good for the patient‘s health, but harmful to the trial.371
Confusion from a logical perspective could compromise a patient‘s care. If a
physician is genuinely unaware of the possible conflicts between research and medicine,
as Miller and Brody argue is often the case in trials, then the physician will be providing
inferior care to his or her patients if he or she has his or her patients participate in such a
trial. If such confusion is as widespread as they contend, the prevalence of medical
research in therapeutic medicine is causing substantial interference with their patients‘
care. Further, confusion on the part of patients as to the sort of craft to which they are
becoming the subject would be a serious blow to the autonomy of patients. Simply by
virtue of their own trust in physicians, patients are not understanding research
methodologies, and this is leading many patients to compromise unwittingly their health
care for the sake of medical research.
However, the final component of Miller and Brody‘s argument does not follow
from a proper distinction between research and medicine. They argue that, if a proper
distinction were made between research and medicine, a distinction between the ethics of
medicine and of research would follow. Miller and Brody hold the belief that ethics of a
370
This would be true even if the motive for picking carrots was becoming healthy. Because the same
activity can be performed for different motives, those motives are not a part of the definition of the activity.
I discussed this issue in Chapter Two, where the Socrates of the Republic makes the distinction between
shepherding, strictly speaking, and the various motives a shepherd might have for shepherding. 371
The compromise of patient care in research trials includes, but is not limited to: the need for placebo
controlled trials for early drug testing; the lack of ―equipoise‖ or equivalence of belief, by a physician as to
whether the test or control arm of a trial is the superior treatment; the inability of patients‘ to consider
internally incommensurable side-effects when choosing a treatment option; the inability to tailor treatments
to a patient‘s medical history; the inability of physicians to customize or to tweak dosage for patients; the
unwillingness of physicians to switch a patient in a trial from one treatment to another unless faces with
serious side-effects; and the use of a ―wash-out‖ period in which patients are untreated (Franklin G. Miller
and Howard Brody, "A Critique of Clinical Equipoise: Therapeutic Misconception and the Ethics of
Clinical Trials", The Hastings Center Report, 2003: 20-1).
184
profession are determined by the nature and purpose of the profession, and with this Plato
would agree. However, it precisely the intrinsic connection between the nature and
purpose of medicine and medical research that requires an intrinsic coneection between
their ethics. Research and medicine are not so neatly divided logically as Miller and
Brody claim. It is true that they have different methods and different immediate goals,
and that these can at times conflict with one another. Nonetheless, they also have a
hierarchical relationship with one another. As I argued in Chapter One, crafts that use
tools and crafts that produce tools are in a hierarchical relationship. What this amounts to
is that the tool-using craft is required to guide the tool-producing craft. The tool-using
craft is required in order to direct the tool-producing craft as to when it should produce its
product. Moreover, tool-producing crafts are incomplete. They only produce tools that
have powers needed by the tool-using craft, and therefore require the tool-using craft to
explain the powers that are needed. Medicine and medical research are in a strict
hierarchy. Medical research by definition produces diagnostic, therapeutic and
prophylactic methods that may be used by medicine in its pursuit of health. In other
words, medical research produces and only produces tools of therapeutic medicine.
Other research on the body, which the Helsinki Declaration refers to as ―theoretical‖, is
explicitly outside of the restrictions of the Helsinki Declaration. As a result, medical
research requires direction from medicine concerning which sorts of procedures need
development and which do not. For example, if a simple, inexpensive cure for a
condition exists, little research is needed for a replacement. However, if a treatment is
costly, only moderately effective or has side-effects, finding a new therapy might be
highly beneficial. Therefore, research is already in a position of needing direction from
medicine for what sorts of treatment are needed. Further, medical researchers depend on
the medical profession to evaluate the outcomes of their various therapies. Whether or
not a particular treatment was efficacious or inefficacious and to what degree is
something that physicians are trained to recognize by virtue of their training as
physicians, and this type of recognition is the very reason that physicians are employed in
research in the first place. They are serving the role of evaluating the various effects of
therapies in terms of various sorts of medical indication and contraindication. In other
words, they are evaluating the products of medical research in terms of the goals of the
185
medical craft. In short, in some ways, current practice already recognizes this Platonic
hierarchy.
This hierarchy should also apply to the ethics. A physician, when involved in
research, isn‘t simply acting as a researcher, where research is some other completely
independent craft from that of medicine. Rather, the physician is involved in the research
precisely in so far as he or she is qualified to evaluate the results, that is, in so far as he or
she is a doctor. This kind of participation is qua doctor, in the sense that one acts as
doctor qua doctor when one is evaluating the efficacy of the medical tools one plans to
use. Take the analogy of a soldier and a shield-maker. The soldier is required to evaluate
the product of a shield-maker. However, when the soldier evaluates the shields, the
soldier is not thereby acting as a shield-maker. It is his or her soldierly knowledge that is
needed to complete the otherwise incomplete craft of shield-making. The soldier does not
become a shield-maker when directing the shield-maker; it is qua soldier that his or her
input is needed. Therefore, Miller and Brody are incorrect in arguing that the different
methods and immediate ends of medical research and medicine imply that they are
independent activities requiring separate ethical codes. In medical research, as with all
tool-producing crafts, tool-users (physicians) are required to evaluate the tools they
employ and are acting as physicians when they do so. They are still acting qua
physicians and do not become researchers when they do so.
Since physicians employed in medical trials are still acting qua physicians, in
much the same way that a soldiers evaluating shields are still acting qua soldiers,
physicians involved in medical trials are still bound by the codes of medical ethics even
though medical research and therapeutic medicine are logically distinct. This means that
physicians employed in research trials are still obliged to follow the ethical obligations of
the medical profession when engaged in medical research, which includes the obligation
to seek optimal care for their patients.372
First, it implies that a physician may not
supervise their patients in a medical trial that would provide those patients with less than
372
Note that this does not contradict Section 2. The optimal care here refers to the care for a patient that
best renders his or her body fit for use in the manner best suited to his or her beliefs about the good life.
186
optimal care.373
This limits researchers in the types of roles that researchers may expect
physicians to play in their research trials. Second, it implies that a physician may not
even refer a patient to a trial that he or she knows will compromise his or her patient‘s
medical care (especially if such a referral is likely to come across as a referral to a
specialist). This implies that researchers must ensure that their trial methodologies
conform to the ethics of the medical profession if they expect to use physicians for any
sort of recruitment of subjects. Medical research ethics must either set itself up without
using physicians qua physicians in their trials (which is likely impossible except possibly
for phase one trials),374
or it must structure its trials in such a way that researchers do not
expect the physicians involved, who are acting as physicians, to contravene the ethics of
their profession.
Is it then possible for there to be any activity of medical research that can exist
independently from medical ethics? It would seem, at first, that there might be the
possibility for there to be pure medical research that does not employ physicians and
therefore would not be bound by ethical restrictions on physicians. However, given the
relationship between a super-ordinate and subordinate craft, any medical research would
necessary be ultimately supervised by medicine.375
So far, I have discussed the role of
physicians in evaluating the results of medical tests. However, physicians are also
required to initiate research. As with meta-analyses, research cannot get started unless
one already has a set of indications and contraindications in place that one is seeking to
promote or avoid. Or, to put it in the terms used in Chapter One, the ―knowing what‖ is
to be produced must be specified by the using craft. As a result, medical research itself is
ultimately commissioned, so to speak, by the medical craft, and cannot exist
373
I do not seek to defend the principle that physicians should always provide their patients with optimal
care here. This principle is a widely-accepted and central principle of most medical codes of ethics and is
accepted as such by Miller and Brody as well. 374
In general, phase one trials test for safety, phase two trials test for absolute efficacy (placebo control) or
relative efficacy (active control) on a small scale, and phase three trials test for relative efficacy on a large
scale. 375
This point is similar to that made by Aristotle in the Nicomachean Ethics: ―...in all of these the ends of
the master arts are to be preferred to all the subordinate ends; for it is for the sake of the former that the
latter are to be pursued‖ (―ἐλ ἁπάζαηο δὲ ηὰ ηλ ἀξρηηεθηνληθλ ηέιε πάλησλ ἐζηὶλ αἱξεηώηεξα ηλ ὑπ‘
αὐηά· ηνύησλ γὰξ ράξηλ θἀθεῖλα δηώθεηαη‖) (1094a14-16). Lower-order crafts are performed for the sake
of those that supervise them. It would be as inappropriate for a physician to undermine the health of his or
her patient in order to develop a new treatment as it would be for a general to undermine a military victory
in order to invent new weapons of war.
187
independently. This relationship of medicine to medical research implies that medical
research must be bound by medical ethics, as physicians should not avoid the strictures of
their professional ethics by employing those who are not themselves bound by them. As
a result, medical research itself, by virtue of its subordinate relationship to medicine,
should be bound by its ethical proscriptions.
4.3 Conclusion
Miller and Brody have discovered an important problem in current research
ethics. There seems to be a great deal of confusion about the possible conflicts between
research and medicine, among physicians and especially among patients. This confusion
stems from a logical confusion between the two activities of research and medicine and
from a failure to appreciate the ways in which these difficulties conflict. However, they
fail to establish that there ought to be two, autonomous codes of ethics for medical
research and for medicine. Medical research and therapeutic medicine are arranged in a
hierarchical, evaluative relationship, and therefore, the physicians involved in medical
research are still involved qua physicians. Therefore, no completely autonomous code of
research ethics is possible.
5 Chapter Conclusion
In this chapter, I have taken the Platonic framework of crafts and applied it to three
modern issues in medicine. The tool of which I have made the most use is Plato‘s
understanding of the hierarchy of crafts. Specifically, crafts are arranged in such a way
that subordinate crafts are incomplete and require a higher-order craft to provide them
with the specifications of their product. The modern paradigm of paternalism, meta-
analyses and concern about the therapeutic misconception treat medicine as too
autonomous, not as a part of a hierarchy of crafts. Once that hierarchy is introduced, new
approaches to these questions take form. In relation to paternalism, the very belief that
medicine is autonomous leads to its own sort of paternalism. As for meta-analyses, the
belief that practical conclusions can come from causal analysis leads to a paternalistic
hybrid of precision and imprecision. As for research ethics, the argument that medicine
188
and research are autonomous fails to take into account that medical research is a
subordinate craft to medicine, and that, as a result, the physicians involved are still
functioning qua physicians, and that they cannot simply be considered to be researchers.
189
C o n c l u s i o n
1 Summary
This dissertation has explored several applications of Plato‘s theory of crafts, and
especially Plato‘s discussion of the craft of medicine. I began in Chapter One with an
overview of Plato‘s theory of crafts. This involved contrasting his use of crafts with
Aristotle‘s definition of the term, as Plato applies the term also to non-productive bodies
of expert knowledge. The inclusion of acquisitive crafts and the acquisitive nature of
theoretical crafts led to a co-extensivity between crafts and “epistēmē‖. I then addressed
some of the features of crafts that recur in the Platonic corpus. First, all crafts have an
important relationship to two meta-crafts, rhetoric and measurement. Second, all crafts
are arranged in a hierarchy in which the tool-user oversees the tool-producer.
This approach to crafts led to numerous puzzles that occupied Plato and his near
contemporaries. It led to the puzzle that when one considers crafts as knowledge, they
seem to be capable of contraries, but when one considers crafts as power, they should be
capable of only one thing. In Chapter Two I discussed Plato and Aristotle‘s approaches
to this difficulty. Both carved out linguistic space in which to provide a substantive
argument to resolve the difficulty. They each have a version of strict speaking and a
metaphysical analysis that prioritizes generation over destruction that allowed them both
to say that, while a person with a craft will be capable of contraries, qua that sort of
craftsperson, they are not. In Chapter Three I discussed an intergenerational debate
between Gorgias and Plato on the nature of freedom. Gorgias had challenged the
traditional synthesis allying freedom with persuasion by arguing that persuasion was a
sort of force. Plato, on the other hand, introduces a new criterion for freedom:
knowledge. I then discussed what he meant by this and showed how Plato arrives at a
conclusion about informing patients that is relevant to modern discussions of informed
consent.
In Chapter Four, I applied Plato‘s approach to crafts to three modern problems.
First, I addressed the problem of paternalism and autonomy. The current paradigm
190
assumes that the physician knows what is best for the patient‘s health, while the patient
can only choose to sacrifice that health for other non-health goals. However, when one
applies Plato‘s understanding of the hierarchy of crafts, one can see that this
understanding is limited and that the goals provided by the patient, who uses the body, is
necessary to complete the medical craft itself. Second, I addressed the problem of
recommendations in medical research. Using data from a recent issue of the Cochrane
Review, I showed how medical researchers systematically confuse practical
recommendations with causal conclusions, and that those who cited those studies
routinely confused the researcher‘s recommendations with the scientific result of the
paper. Finally, I looked at the ―therapeutic misconception‖ discussed by Miller and
Brody and argued that, while they were correct that physicians and patients often confuse
the ends of research and medicine, their claim that this implied that research and
medicine should have different ethical codes failed to account for the hierarchy between
medical research and medicine proper. The richness and unity of Plato‘s framework can
be appreciated by the way that it can be applied to so many subjects in the medical field.
2 Important Results
One important result of this dissertation was to bring together the various uses to
which Plato puts medicine in his corpus. While I did not evaluate all of his references, I
chose a representative sample. Plato uses medicine as his paradigmatic craft, and turns
again and again to this example in almost every dialogue. Medicine serves him well as
an example because:
1. It has a definite product: health.
2. It has a definite subject: the body.
3. It is a craft and therefore a body of knowledge.
4. Its subject is made well through the application of mathematical ratios.
Medicine has all of the characteristics that makes it the ideal example of a craft. This has
two effects. First, almost everything that Plato says about medicine should be taken as a
part of a general theory of crafts. Plato‘s discussion of medicine brings to light an entire
approach of the relationship between knowledge, power and practical wisdom. Second,
and conversely, Plato almost inadvertently supplies an entire approach to medical ethics.
191
Though he almost always uses medicine as an analogy for something else, he ends up
saying so much about medicine that one uncovers a Platonic approach to medicine as a
whole. After all, as discussed in the introduction, Plato mentions medicine than he does
politics, and that he isn‘t ultimately interested in medicine doesn‘t imply that he said any
less about it. That there has not been a great deal of interest in Plato‘s approach to
medicine reflects Plato‘s own ultimate interests, but we are under no obligation to share
only Plato‘s ultimate interests when studying his works.
Another major goal in this dissertation was to examine the ways in which Plato‘s
understanding of crafts could be applied to the medical and professional ethics. The
ethics of professions, especially that of medicine, are themselves intellectually
challenging and morally problematic. Many of the dilemmas within those professions,
even if they cannot be solved by Plato‘s understanding of the crafts alone, can greatly
benefit from an application of Plato‘s understanding of the crafts. First, using Plato can
help resolve definitional questions about the limitations of a given field. For instance,
Plato‘s understanding of crafts can be used to establish where medicine ends and research
begins, or where being a lawyer ends and making money begins. Just as Socrates used
the ―strict speaking‖ sense of a craft in Republic I to establish that tyrants aren‘t strictly
speaking rulers, so too one can use the ―strict speaking‖ sense of craft to clearly define
what sorts of activity lie within the scope of a profession. Second, using Plato can help
establish the kinds of hierarchical relationships that ought to be in place between these
professions once they are delineated. When research ethics dictates medical ethics or
when weapons manufacturers dictate military policy, the quality of medical treatment and
military strategy suffers. These hierarchical relationships help professionals to appreciate
the different roles that they play relative to other professions, including those of
supervision, tool-production and co-operation.
3 Future Applications
The topics to which I applied Plato‘s theory of crafts are only some of the topics
to which Plato‘s theory of crafts has application. There are many other topics that there
was not space to cover. In this conclusion, I will suggest further areas in modern
192
biomedical ethics to which the framework would be useful. These subjects will be an
important component in my future research.
First, Plato‘s framework of health can be very helpful in understanding the nature
of medical distribution questions at both the institutional and the political levels. Norman
Daniels argued in Just Health Care that health care is a matter of justice because of the
kind of opportunities (powers) that it gives to those who are healthy, while Ezekiel
Emmanuel discussed in The Ends of Human Life: Medical Ethics in a Liberal Polity the
ways in which various kinds of health care distribution result from different views of the
good life. Distribution of health care can both augment and diminish autonomy, and a
study of those ways with reference to Plato‘s understanding of the hierarchy between
medicine and the body-using patient would be a worthwhile project.
Second, Plato‘s framework raises some important questions for the kind of health
education that is important for free citizens. Although they may occur only once or twice
in our lives, important medical decisions are often decisions that can affect our capacity
to live the kind of life we wish. Having the skills to consider health care alternatives in
light of the various life plans we might have is an important part of ensuring that our
autonomy is not compromised when we enter a health-care context. Understanding
basically how our bodies function can provide students with the background they need to
consider health care decisions autonomously. Much of our current health education
focuses on instilling rules (like the four food groups) rather than understanding. What
kind of information or skills a free citizen needs to be an autonomous before entering a
clinical setting is a topic rarely addressed and about which Plato‘s framework has much
to offer. His framework challenges us to consider the kind of knowledge we ourselves
need as autonomous patients.
Third, in the Gorgias, Socrates makes a claim that he makes in no other dialogue.
He claims that ―gymnastikē‖ or gymnastics, the craft of staying healthy, and ―iatrikē‖ or
medicine, the craft of restoring health, are the same craft, because they have the same
product, health. This claim mirrors the modern distinction between ―preventative‖ and
―curative‖ medicine, the former of which is normally considered preferable. Some
interesting puzzles arise, however, from expanding the definition of ―medicine‖ beyond
that of restoring health. Once one includes prevention, odd conclusions seem to follow,
193
like the inclusion of sanitation and air-quality control. Even more puzzling would be the
social determinates of health, which seem to imply that poor people are often unhealthy
just by being poor, and that improving social equality might therefore be a part of the
same craft of medicine. What we call ―medicine‖ may find itself in a hierarchical or part-
whole relationship to a broader, health-producing craft. How to fit together all of these
various health-producing crafts in using a Platonic framework is another project to
consider.
These are just three possible applications. The Platonic framework of crafts is so
rich that it could help contribute to numerous questions in the field of medical ethics,
many of which I have not discussed myself. Moreover, though this is beyond the scope
of the present thesis, Plato‘s framework may well be applicable to other areas of
professional ethics. Plato‘s approach to crafts provides the reader with a framework in
which to consider the relationships between knowledge, power and practical wisdom.
194
B i b l i o g r a p h y
Aristotle. De Partibus Animalium I and De Generatione Animalium I. Edited by David
M. Balme. New York: Oxford University Press, 1992.
—. Metaphysics, Books [Gamma], [Delta] and [Epsilon]. Edited by Christopher Kirwan.
Oxford: Clarendon Press, 1971.
—. Metaphysics: Book [Theta]. Translated by Stephen Makin. New York: Oxford
University Press, 2006.
—. On the Parts of Animals I-IV. Translated by James G. Lennox. New York: Oxford
University Press, 2001.
—. The Complete Works of Aristotle: the Revised Oxford Translation. Edited by Jonathan
Barnes. Princeton: Princeton University Press, 1984.
Arpaly, Nomy. "Which Autonomy?" In Freedom and Determinism (Topics in
Contemporary Philosophy), by Joseph Kleim Campbell, Michael O'Rourke and
David Shier, 173-188. Cambridge: The MIT Press, 2004.
Bäck, Allan T. Aristotle's Theory of Predication. Boston: Brill, 2000.
Bader, Daniel. "Using Plato‘s and Aristotle‘s Craft Analogies to Understand Biological
Relationships." Examining Telelogy Conference, March 26, 2010.
Balansard, Anne. Technè dans les Dialogues de Platon: L’empreinte de la sophistique.
Sankt Augustin: Academia-Verl., 2001.
Barney, Rachel. "Gorgias' Defense, Plato and His Opponents on Rhetoric and the
Good." The Southern Journal of Philosophy 48, no. 1 (2010): 95-121.
Benchimol, E. I., C. H. Seow, A. R. Otley, and A. H. Steinhart. "Budesonide
Formaintenance of Remission in Crohn‘s Disease." Cochrane Database of
Systematic Reviews, 2009: CD002913.
Berg, Jessica W., Paul S., Parker, Lisa S. Appelbaum, and Charles W. Lidz. Informed
Consent: Legal Theory and Clinical Practice. New York: Oxford University
Press, 1987.
Bobonich, Christopher. Plato's Utopia Recast. Oxford: Oxford University Press, 2002.
Boorse, Christopher. "On the Distinction Between Disease and Illness." Philosophy and
Public Affairs 5 (1975): 49-69.
Briones, E., J. R. Lacalle, and I. Marin. "Transmyocardial Laser Revascularization
Versus Medical Therapy for Refractory Angina." Cochrane Database of
Systematic Reviews, 2009: CD003712.
Carlevale. "Education, "Phusis," and Freedom in Sophocles' "Philoctetes"." Arion, Third
Series 8, no. 1 (Spring-Summer 2000): 26-60.
Catan, Giovanni Reale and John R. The Concept of First Philosophy and the Unity of the
Metaphysics of Aristotle. Albany: State University of New York Press, 1980.
Caughey, A. B., J. M. Nicholson, and A. E. Washington. "First- vs Second-Trimester
Ultrasound: The Effect on Pregnancy Dating and Perinatal Outcomes." American
Journal of Obstetrics and Gynecology, no. 198 (2008): 703.e1-703.e6.
Caughey, A. B., N. E. Stotland, Washington A. E., and G. J. Escobar. "Who is at Risk for
Prolonged and Postterm Pregnancy?" American Journal of Obstetrics and
195
Gynecology, no. 200 (2009): 683.e1-683.e5.
Consigny, Scott. Gorgias: Sophist and Artist. Columbia: University of South Carolina
Press, 2001.
Costa, Maria L., Jose G. Cecatti, Helaine, M. Milanez, Joao P. Souza, and Metin
Gülmezoglu. "Audit and Feedback: Effects on Professional Obstetrical Practice
and Healthcare Outcomes in a University Hospital." Acta Obstetricia et
Gynecologica Scandinavica 88, no. 7 (2009): 793-800.
Daniels, Norman. "Health Care Needs and Distributive Justice." Philosophy and Public
Affairs 10, no. 2 (1981): 146-179.
—. Just Health Care. Cambridge: Cambridge University Press, 1985.
Davies, Paul Sheldon. Norms of Nature: Naturalism and the Nature of Functions.
Cambridge: MIT Press, 2003.
Dorter, Kenneth. "Philosopher-Rulers: How Contemplation Becomes Action." Ancient
Philosophy 21 (2001): 335-356.
—. The Transformation of Plato's Republic. Oxford: Lexington Books, 2006.
Emmanuel, Ezekiel. The Ends of Human Life: Medical Ethics in a Liberal Polity.
Cambridge: Harvard University Press, 1991.
Euripides. Heraclidae. Edited by John Wilkins. Oxford: Clarendon Press, 1979.
Euripides. The Plays of Euripides. Vol. 5, in Great Books of the Western World, edited by
Mortimer J. Adler, translated by Edward P. Coleridge, 203-454. Chicago:
Encyclopædia Brittanica, 1952.
Franklin, Lee. "Technē and Teleology in Plato‘s Gorgias." Apeiron 38, no. 4 (2005): 229-
255.
Gatward, H., M. Simpson, L. Woodhart, and M. Stainton. "Women‘s Experiences of
Being Induced for Post-Date Pregnancy." Women and Birth 23, no. 1 (2009): 3.
Geach, Peter. "Good and Evil." In Theories of Ethics, edited by Philippa Foot, 64-73.
New York: Oxford University Press, 1967.
Gülmezoglu, A. M., C. A. Crowther, and P. Middleton. "Induction of Labour for
Improving Birth Outcomes for Women at or Beyond Term." Cochrane Database
of Systematic Reviews, 2006: CD004945.
Halliday, H. L., and R. A., Doyle, L. W. Ehrenkranz. "Early (< 8 days) Postnatal
Corticosteroids for Preventing Chronic Lung Disease in Preterm Infants."
Cochrane Database of Systematic Reviews, 2009: CD001146.
Halliday, H. L., R. A. Ehrenkranz, and Doyle L. W. "Late (>7 Days) Postnatal
Corticosteroids for Chronic Lung Disease in Preterm Infants." Cochrane
Database of Systematic Reviews, 2009: CD001145.
Heimstad, Runa, Pål R. Romunstad, and Kjell Å. Salvesen. "Induction of Labour for
Post-Term Pregnancy and Risk Estimates for Intrauterine and Perinatal Death."
Acta Obstetricia et Gynecologica Scandinavica 87, no. 2 (2007): 247-249.
Hesiod. "Works and Days." In The Homeric Hymns and Homerica with an English
Translation, edited by Hugh G. Evelyn-Wight, 45-65. London: Harvard
University Press, 1914.
Hippocrates. Hippocratic Writings. Edited by G. E. R. Lloyd. Middlesex: Penguin Books
Ltd., 1978.
Hornblower, Simon, and Anthony Spawforth. The Oxford Companion to Classical
Civilization. Oxford: Oxford University Press, 2004.
196
Hurka, Thomas. Virtue, Vice and Value. New York: Oxford University Press, 2001.
Irwin, Terence. Aristotle's First Principles. New York: Oxford University Press, 1988.
—. Plato's Ethics. Oxford: Oxford University Press, 1995.
Isocrates. Nicocles or the Cyprians. Vol. 1, in Isocrates With an English Translation in
Three Volumes, edited by George Nolin, 74-115. London: William Heinemann
Ltd., 1980.
Janaway, Christopher. Images of Excellence: Plato's Critique of the Arts. Oxford: Oxford
University Press, 1999.
Josephus. The Antiquities of the Jews, Books 18-19. Translated by Louis H. Feldman.
Cambridge: Harvard University Press, 1965.
Jouanna, Jacques. Hippocrates. Baltimore: Johns Hopkins University Press, 1999.
Khan, J., and S. A. Razak. "Preoperative Nutrition to Prevent Post Operative
Complications in Malnourished Patients Undergoing Elective Abdominal
Surgery." Cochrane Database of Systematic Reviews, 2009: CD007597.
Knight, Brenda. Women of the Beat Generation: The Writers, Artists and Muses at the
Heart of a Generation. Boston: Conari Press, 1996.
Laks, André. "Freedom, Liberty and Liberality in Plato's Laws." Social Philosophy and
Policy 24, no. 2 (Summer 2007): 130-152.
Levy, David. "Technē and the Problem of Socratic Philosophy in the Gorgias." Apeiron
38, no. 4 (2005): 185-228.
Liddell, H.G., and R. Scott. Greek-English Lexicon with a Revised Supplement. Oxford:
Clarendon Press, 1996.
Lloyd, G. E. R. Methods and Problems in Greek Science. Cambridge: Cambridge
University Press, 1991.
Lyons, John. Structural Semantics: An Analysis of Part of the Vocabulary of Plato.
Oxford: Basil Blackwell, 1963.
MacIntyre, Alasdair. After Virtue. Third. South Bend: University of Notre Dame Press,
2007.
McGuinness, B., R. Bullock, and D., Kerr, E., Passmore, P. Craig. "Statins For the
Treatment of Alzheimer‘s Disease and Dementia." Cochrane Database of
Systematic Reviews, 2009: CD007514.
Miller, Franklin G., and Howard Brody. "A Critique of Clinical Equipoise: Therapeutic
Misconception and the Ethics of Clinical Trials." The Hastings Center Report,
2003: 19-28.
Mozurkewich, E., J. Chilimigras, E. Koepke, K. Keeton, and V. King. "Indications for
Induction of Labour: A Best-Evidence Review." British Journal of Obstetrics and
Gynecology, no. 116 (2009): 626-636.
Nicholson, J. M., M. H. Stenson, L. C. Kellar, A. B. Caughey, and G. A. Macones.
"Active Management of Risk in Nulliparous Pregnancy at Term: Association
Between a Higher Preventive Labor Induction Rate and Improved Birth
Outcomes." American Journal of Obstetrics and Gynecology, no. 200 (2009):
254e1-254e13.
Nys, Thomas, Yvonne Denier, and Toon Vandevelde. Autonomy & Paternalism:
Reflection on the Theory and Practice of Health Care. Leuven: Peters, 2007.
Olsen, O., Middleton, P., Ezzo, J., P., Hadhazy, V. Gotsche, A. Herxheimer, J. Kleijnen,
and H. McIntosh. "Quality of Cochrane Reviews: Assessment of Sample from
197
1998." British Medical Journal, no. 323 (2001): 829-832.
Oxford English Dictionary. "free, adj., n., and adv." OED Online. March 2010.
http://dictionary.oed.com/cgi/entry/50089637 (accessed June 1, 2009).
—. "jugglery" OED Online. March 2010. http://dictionary.oed.com/cgi/entry/50124579
(accessed September 2, 2010).
Parry, Richard D. Plato's Craft of Justice. Albany: State University of New York Press,
1996.
Pascual, J., J. Zamora, C. Galeano, A. Royuela, and C. Querada. "Steroid Avoidance or
Withdrawal for Kidney Transplant Recipients." Cochrane Database of Systematic
Reviews, 2009: CD005632.
Pileggi, Nicholas. Wiseguy: Life in a Mafia Family. New York: Pocket Books, 1987.
Plato. Euthydemus. Edited by Rosamond Kent Sprague. New York: Hackett, 1993.
—. Laws X: Translated With a Text and Commentary. Translated by Robert Mayhew.
Oxford: Clarendon Press, 2008.
—. Philebus. Edited by Dorothea Frede. Indianapolis: Hackett, 1993.
—. Plato: Complete Works. Edited by John Cooper. Indianapolis: Hackett, 1997.
Roberts, Royston M. Serendipity: Accidental Discoveries in Science. Toronto: John
Wiley & Sons, 1989.
Robinson, Eric W. The First Democracies: Early Popular Government Outside Athens.
Stuttgart: Steiner, 1997.
Rogers, L., Siu S. S. Nelson, D. Luesley, and H. O. Dickinson. "Adjuvant Radiotherapy
and Chemoradiation after Surgery for Cervical Cancer." Cochrane Database of
Systematic Reviews, 2009: CD007583.
Roochnik, David. "Socrates' Use of the Techne-Analogy." Journal of the History of
Philosophy 24 (July 1986): 295-310.
Ryle, Gilbert. "Knowing How and Knowing That." Proceedings of the Aristotelian
Society 46 (1946): 1-16.
Sprague, Rosamond Kent, ed. The Older Sophists. Indianapolis: Hackett, 1973.
Stotland, N. E., A. E. Washington, and A. B. Caughey. "Prepregnancy Body Mass Index
and the Length of Gestation at Term." American Journal of Obstetrics and
Gynecology, no. 197 (2007): 378.e1-378.e5.
Taylor, F., K. Ward, T. H. M. Moore, M. Burke, G. Davey Smith, and S. Ebrahim.
"Statins for the Primary Prevention of Cardiovascular Disease." Cochrane
Database of Systematic Reviews, 2009: CD004816.
Thucydides. The History of the Peloponnesian War, Books 1-2. Translated by C. F.
Smith. London: Heinemann, 1919.
Tiles, J. E. "Technē and Moral Expertise." Philosophy: The Journal of the Royal Institute
of Philosophy, no. 59 (1984): 49-66.
Todd, S. C. A Commentary on Lysias: Speeches 1-11. Oxford: Oxford University Press,
2007.
University of California. Thesaurus Linguae Graecae. March 18, 2009.
http://www.tlg.uci.edu/ (accessed September 22, 2009).
Vellekoop, J, F. P. J. M. Vrouenraets, J. W. van der Steeg, B. W. J. Mol, and F. J. M. E.
Roumen. "Indications and Results of Labour Induction in Nulliparous Women:
An Interview Among Obstetricians, Residents and Clinical Midwives." European
Journal of Obstetrics and Gynecology 146, no. 2 (2009): 156-159.
198
Wachbroit, Robert. "Normality as a Biological Concept." Philosophy of Science, no. 61
(1994): 579-591.
Walters, J. A. E., P. G. Gibson, R. Wood-Baker, M. Hannay, and E. H. Walters.
"Systemic Corticosteroids for Acute Exacerbations of Chronic Obstructive
Pulmonary Disease." Cochrane Database of Systematic Reviews, 2009:
CD001288.
Watanabe, N., R. Churchill, and T. A. Furukawa. "Combined Psychotherapy Plus
Benzodiazepines for Panic Disorder." Cochrane Database of Systematic Reviews,
2009: CD005335.
Wennerholm, Ulla-Britt, Henrik Hagberg, Bengt Brorsson, and Christina Bergh.
"Induction of Labor Versus Expectant Management for Post-Date Pregnancy: Is
there Sufficient Evidence for a Change in Clinical Practice?" Acta Obstetrica et
Gynecologica Scandinavica 88, no. 1 (2009): 6-17.
Witt, Charlotte. Substance and Essence in Aristotle: An Interpretation of Metaphysics
VII-IX. Ithaca: Cornell University Press, 1994.
Woolf, Raphael. "Why is Rhetoric Not a Skill?" History of Philosophy Quarterly 21, no.
2 (April 2004): 119-130.
World Medical Association. "Declaration of Geneva." Australian Medical Association
Website. May 2005. http://www.ama.com.au/node/2474 (accessed April 14,
2010).
—. "World Medical Association Declaration of Helsinki: Ethical Principles for Medical
Research Involving Human Subjects." World Medical Association Website.
October 20, 2008.
http://www.wma.net/en/30publications/10policies/b3/index.html (accessed
November 4, 2009).